Rev 3.1.13 Structured abstract {2149,1250 words} Background: Suicide rates between 1999 and 2016 have been on the rise in the United States Examining state-level trends in, and contributing circumstances to, suicide, can inform movement towards the goal of comprehensive state suicide prevention. Methods: Trends in age?adjusted suicide rates, overall and by sex and state, among people aged :10, were calculated using data from the National Vital Statistics System. Changes in rates and state rankings were assessed across six consecutive th ree-vear periods [1999-2001, 2002-2004, 2005-2002, 2008-2010, 2011-2013, 2014?2016}. Data from the 2015 National Violent Death Reporting System, across 2? states, were analyzed to compare circumstances [aetween suicide decedents with and without known mental health problems Results: Average annual percentage change in suicide rates across the period, increased significantlilr in the U.S. overall and in 44 states, ranging from 5.996% in Delaware to 52.6% in North Dakota. People with and without known mental health problems, experienced a range of multiple contributing circumstances to their suicides, including substance use, relationship problems, and recent crises. Conclusions: Suicide rates have risen significantlv in the 11.5. and across most states from 1999-2016. [\lumerous factors contribute to sulcide among people with and without known mental health problems. Implications for Public Health Practice: To reverse upward trends in suicide, a population-based public health approach inclusive of evidence?based strategies across multiple levels (individual, familw'relationship, communitv, societal}, focused on preventing risk before it starts, identifving and supporting people alreadv at risk. preventing-reattempts, and caring for survivors after a suicide, is needed. INTRODUCTION words} In 2016, nearly 45,000 suicides occurred in the United States While overall rates have been on the decline globallv;2 rates ofsuicide in the 0.5. have increased between 1999 and 2015 across most age groups, among males and females, across racialj?ethnic groups, and across urbanization levels.3 ??ates of nonfatal attempts have also been on the rise, with emergencv department visits for self?harm injurv having increased more than 40% between 2001 and 2015.5 in er, totaled more than $69 billion in direct medical and work loss costs. Suicide is rarely caused bv one thing; rath the risk factors for suicide are numerous and occur at multiple levels (individual, familvfrelationship, communitv, and societal}? However, Ehe focLis tends to point towards mental illnesseskuch as depression and i Comment Does this seem too abrupt or unclear as in, ?why look at suicide within those two groups?" Comment Couid also sav, no single factor alone contributes to suicide. Comment if NCHS data brief has urbanization we can take this reference out if we need space for others. 1 Comment This needs updating for 2016 but ED data is still for 2015. r' 1 i if Comment Need idea for 3 lil[ reference here. 4i 'u bipolar disorder, as the cause. Other factors associated with suicide include lack of social connectedness, access to quality evidencebased treatment, irresponsible media reporting, lack of skills, stigma associated with help-seeking, and a prior suicide attempt, among others. While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, the U.S. continues to struggle to make this a realitv due to a lack of funding for implementation of evidencevbased prevention strategies, misperceptions about suicide preventabilitv, and a prevention focus primarilv at the individual level. 1 Wisqars '1 World raport 3 NCHS Data brief 2018 Kegler et al 2017:1 5 WISQARS Nonfatal injuryI reports 5 National Strategy for Suicide Prevention Rev 3.1.13 METHODS [1191250 words} Suicide rate estimates and trend analyses exclude data for persons <10 yea rs old. Age?specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {lnternationai Classification of Diseases 10'? Revision underlying?cause-of death codes KEG-X34, YBIO, *U03l. Age-specific population estimates were obtained from Ll.5. Census Bureau National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2016. Rate estimates were age-adjusted to the 0.5 year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from the 2015 National 'v'iolent Death Reporting System from 2? states, were used to compare the characteristics, including precipitating circumstances, ofsuicide decedents with and without known mental health problems. compiles data from three primary data sources: death certificates, coronerfmedical examiner reports {including toxicology}, and law enforcement reports. RESULTS usersoo words] The most recent overall suicide rates (representing 2014-2015] ranged from 6.9 (District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference {Table Across the entire study period, rates increased in all but one state {Nevada}, with increases ranging from +0.2 [Delaware] to +8.1 {Wyoming}. Percentage increases in rates ranged from +5.50% {Delaware} to (North Dakota}, with percentage increases of at least 25% observed in over half of all states as well as nationally. beographically, many states showing the largest percentage increases are in the upper Great Plains and the upper lntermountain West Modeled suicide rates trends were found to be significantly increasing for44 states, as well as for the U5. overall [Table By sex, rate trends were found to be significantly increasing in 34 states for males and in 43 states for females. Nationally, the model?estimated AAPC for overall suicide rates was By sex, the national was +11% for males and +16% for females. INSERT RESULTS HERE Eilscussmiv {east/run words] The current study highlights state?specific trends in suicide among people a 10 years. over time by state and by sex. It also provides a snapshot of circumstances of suicide among people with and without known mental health problems to provide insight into the growing suicide problem in the US. Findings that national suicide rates increased significantly, overall and among both males and females, aligns with past studies reporting on rate increases over time and across demographic subgroups?r?i-7 To extend this information, novel state-specific trend analyses identified overall suicide rate increases in 44 states. Five remaining states and DJ: reported non-significant rate increases and Nevada, showed an overall decline in suicide rates. The geographic distribution of rate increases suggests that many regions already known to have the highest rates of suicide in the country leg. mountain west}, also experienced some of the largest percent increases in suicide rates. Conversely, regions that once seemed more protected from suicide northeast, midwest] also Asha?s paper ., [Comment We need to consider changing this as I don?t have space to 5 include in the abstract, intro, and it hasn't i been included in the methods?basically i ,not incorporated at all. ,i Comment We also have some additional data that is forthcoming on the types of recent crises, FYI so that may be hadded as well. Comment Thinking to put something in here about the overlap in 5. JL suicide rates and opioid overdose deaths Rev 3.1.13 epidemic that occurred over the same time period,EL and the downturn in the ECONOMY in the period 20m.2nn9,9 May take this . . . 1 i I ome studies also suggest a negative Impact of social media, pa rtlcularly among youth. {Comment [5111]: Is this too pointed or strong? ,i Comment ISDU: This is a Related to gender, female suicide rates increased in more states {43] than did male suicide rates This may ifferent way to present this info. What hint at a narrowing of the suicide gender gap, as has been found related to suicide in other demographic groupings by urbanization It also aligns with reports of large increase in middle aged female suicides between 1999-2014.? This may signal an increases in risk factors more common among females leg, 5" intimate partner violence] or may reflect a differential impact of common suicide risk factors substance 3i i u' do you think? Comment Is it nationally representative at this point? Or no? abuse) by sex. More research on this topic is needed. Mental illness is an import risk factor for suicide. Unlike other studies,12 the current study found that more than ,l half of people did not have a known mental health problem. Among people who did, just over half were known to be in treatment at the time of their deaths, about one?third were reported to never have received any _5 mental health treatment, and between about 30-40% had a history of ideation andfor suicide attempts. This i points to the need for increased identification of, and improved responses to, people at risk, and highlights the need for better access to care that is evidence-based collaborative}, affordable, and sustained in order to 5' prevent suicide attempts, re-attempts, and suicide. i .1 . . While there is a tendency to focus primary attention to the risk fact of mental illness, risk Factors for suicide are i if i allowed for the examination of other contributing risk factors within these groups. Results suggest many differences in the prevalence of circumstances, for example people without mental health problems were more likely to have criminal?legal problems, evictionfloss of home, and recent crises. However, some circumstances such as intimate partner problems, argumentsx'conflicts, recent crises, and physical health problems are common across groups. Whether the two groups are viewed as more similar or more unique, results are clear that multiple challenging life circumstances impacted both. Both state trends and circumstance findings point to the same solution?the need for a comprehensive public health approach to suicide, inclusive of multiple and multi-level evidence-based strategies and approaches that' i} prevent suicide risk before it occurs, by strengthening economic supports, strengthening access to, and coverage of, care that is safe and effective; creating protective environments ie.g. reducing access to lethal means among people at risk); promoting connectedness, and teaching coping and problem-solving skills; ii] identify and support people already at risk, e.g. through gatekeeper training, crisis intervention. referral to ,1 effective treatments, and prevention of re-attempts; and lessen harms after a suicide ie.g., supporting i i survivors and promoting responsible media reporting to avoid suicide ,i able to implement such a comprehensive approach. Limitations The findings provided of this report are subject to at least two limitations. Data for state level analyses extend to 2016, the date for which the most recent mortality data is available, and ciata from come from 2015 and cover only 2? statei?bstractors of data are they 5 Rudd et al MMWH 2015 Need ref here 1? Need ref '12 13 Stone DM. Holland KM, Bartholow EN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: 3 technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 201?. Rev 3.1.13 receive. For example, medical and mental health information are not captured directly from medical records but from coronerfmedical examiner reports and the decedent?s family members and friends. Therefore, the completeness and accuracy of this information are limited by the knowledge of the informant. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies.14 Conclusions Suicide is a growing public health problem with states experiencing large increases in rates between 1999 and 2016. Mental illness is an important risk factor for suicide and it is one of many factors. Comprehensive suicide prevention strategies are needed in states that encompass risk factors at multiple levels and that work to prevent suicide risk before it occu rs, to identify and successfully support people at?risk, and that lessens future harms. Corresponding author References Tables and figures 1? Need ref here Rev 3.2.13 abrupt or unclear as in, ?why look at [Comment Does this seem too suicide within those two groups?" Structured abstract {2149,1250 words} Background: Suicide rates between 1999 and 2016 have been on the rise in the United States Examining state-level trends in, and contributing circumstances to, suicide, can inform movement towards the goal of i {Comment Couid also sav, no single factor alone contributes to suicide. Comment ism]: rr NCHS data brief has urbanization we can take this reference out if we need space for others. comprehensive state suicide prevention. Methods: Trends in age?adjusted suicide rates, overall and by sex and state, among people aged :10, were calculated using data from the National Vital Statistics System. Changes in rates and state rankings were assessed across six consecutive th ree-vear periods [1999-2001, 2002-2004, 2005-200}, 2008-2010, 2011-2013, ,3 .i . 2014?2016}. Data from the 2015 National Violent Death Reporting System, across 2? states, were analyzed to quammeni This needs updating far1 compare circumstances [aetween suicide decedents with and without known mental health problems] if :5 2016 but ED data is still for 2015. {f ff?mment Need idea for 1 if: reference here. 1 opioids as an example here instead so Results: Average annual percentage change in suicide rates across the period, increased in the U5. that we can tie this in?? overall and in 44 states, ranging from 5.996% in Delaware to 516% in North Dakota. People with and without known mental health problems, experienced a range of multiple contributing circumstances i. Comment we want their suicides, including substance use, relationship problems, and recent crises. Conclusions: Suicide rates have risen significantlv in the LLB. and across most states from 1999-2016. [\lumerous factors contribute to suicide among people with and without known mental health problems. Implications for Public Health Practice: To reverse upward trends in suicide, a population-based public health approach inclusive of evidence?based strategies across multiple levels (individual, familw'relationship, communitv, societal}, focused on preventing risk before it starts, identifving and supporting people alreadv at risk. preventing-reattempts, and caring for survivors after a suicide, is needed. INTRODUCTION words} In 2016, nearly 45,000 suicides [13.4f100,000i occurred in the United States While overall rates have been on the decline globallv;2 rates ofsuicide in the US. have increased between 1999 and 2015 across most age groups, among males and females, across raciali?ethnic groups, and across urbanization levels.3 ?itatespfuwj nonfatal attempts have also been on the rise, with emergencv department visits for self?harm injurv having increased more than 40% between 2001 and 2015.5 in totaled more than $69 billion in direct medical and work loss casts. Suicide is rarely caused bv one thing,- rather, the risk factors for suicide are numerous and occur at muitiple levels (individual, familvfrelationship, communitv, and societal}? However, Ehe focUs tends to point towards mental illnesseskuch as depression and i. bipolar disorder, as the cause. Other factors associated with suicide include lack of social connectedness, access to quality evidencebased treatment, irresponsible media reporting, lack of skills, stigma associated with help-seeking, and a prior suicide attempt, among others. While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, the U.S. continues to struggle to make this a realitv due to a lack of funding for implementation of evidencevbased prevention strategies, misperceptions about suicide preventabilitv, and a prevention focus primarilv The 1 Wisqars '1 World report 3 NCHS Data brief 2018 Kegler et al 201? 5 WISQARS Nonfatal injuryI reports 5 National Strategy for Suicide Prevention Rev 3.2.13 current study examines increases in state suicide rates over time and circumstances contributing to suicide among people with and without mental health problems. METHODS (119/250 words} Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Statistics System coded death certificate records ilnternotionoi Classi?cation of Diseases 10?h Revision underlying-causEvof death codes l031.0, *003}. Age-speci?c population estimates were obtained from 0.5. Census Bureau National Center for Health Statistics bridged?race population i data releases. National and state?level suicide rate estimates were calculated for six consecutive three?year aggregate periods covering years 19994016. Rate estimates were age?adjusted to the 0.3. year 2000 standard population and expressed per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from the 2015 National Violent Death Reporting System from 27 states, were used to compare health problems. compiles data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology}. and law enforcement reports. RESULTS (2561500 words} The most recent overall suicide rates (representing 2014?2015} ranged from 6.9 (District of Columbia {0.01} to 29.2 (Montana} per 100,000 persons per year. a four?fold difference (Table Across the entire study period. rates increased in all but one state {Nevada}, with increases ranging from +0.2 (Delaware) to +8.1 {Wyoming}. Percentage increases in rates ranged from +59% {Delaware} to +57.6% (North Dakota}, with percentage increases of at least 25% observed in over half of all states as well as nationally. beographically, many i 1 states showing the largest percentage increases are in the upper Great Plains and the upper lntermountain West ooh-q..- Modeled suicide rates trends were found to be significantly increasing for 44 states, as well as for the US. overall (Table By sex, rate trends were found to be significantly increasing in 34 states for males and in 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex. the national AAPC was for males and +25% for females. Suicide decedents with and without known mental health problems were compared. Both groups were predominately male and predominately non?Hispanic white. Fifteen percent of those with known mental health problems and 20% without had ever served in the U.S. military. Suicide decedents without known MH problems had 2.3 greater odds of being male (95% Confidence interval [Cl] and significantly greater odds of racei?ethnicity other than non?Hispanic white (odds ratio range: 1.0-2.1; 95% Cl range [1.04.3] [1010]}. They had significantly greater odds of dying by suicide after committing homicide(s} (adjusted odds ratio [aDFt] 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2. 95% CI Although firearms were the most common injury mechanism for both groups, suicide decedents with known I'v'il-i problems more often died by poisoning than those without MH problems (19.0% vs. most frequently due to overdose of over?the-counterfotherwise unclassified drugs opioids antidepressants or benzodiazepines Comment Need to decide on decedent or victims. Both are used here. I refer decedent. Comment We need to consider changing this as I don?t have space to include in the abstract, intro, and it hasn?t been included in the methods?basically not incorporated at all. All suicide decedents with known NIH problems and approximately 85% without known MH problems additional data that is forthcoming on the types of recent crises, FYI so that may be added as well. {Comment Thinking to put Rev 3.2.13 [Comment We also have some (11:93:52] had known precipitating circumstances. Decedents without known MH problems had significantly greater odds of any type of relationship problem or loss laOH 1.3, 95% CI intimate partner problems specifically laOR 1.4, 95% CI an argument or conflict 1.4, 95% CI and having been a recent perpetrator of interpersonal violence laDFi 2.0, 95% CI 1.52.4} noted as a factor precipitating suicide. Decedents with known MH problems more frequently had a history of mental health or substance abuse treatment [62.2% vs. and co-occurring substance abuse problems {31.6% vs. 25%, 5 .01), something in here about the overlap in suicide rates and opioid overdose deaths. Comment May take this out. Suicide decedents without known MH problems had significantly greater odds of other life stressors as precipitating circumstances, including recentfimpending criminal legal problems 1.7, 95% CI and eviction/loss of home 1.4, 95% CI They had significantly lower odds of recent release from an institution of any kind, but when this was indicated, significantly greater odds of recent release from a correctional facility 4.5, 95% CI or hospital 1.3, 95% CI Of those decedents with known NIH problems who were recently released from an institution, 42.8% had been recently released from a institution. Those with known l'v?lH problems also more frequently hadjob or financial problems {16.8% vs. 15.6%; 5 .05). Decedents without known MH problems had significantly greater odds of a recentfimpending crisis 1.4, 95% CI When the type ofcrisis was known, it was most frequently a problem related to an intimate ,i partner physical health criminal legal issues a family relationship or a job i Over one-fourth of decedents with a known MH problem also had recent or impending crises, most If frequently related to problems with an intimate partner physical health or a family relationship i Decedents without known MH problems had significantly greater odds of criminal legal faUFt 1.6, 95% CI and significantly lesser odds of job-related 0.195% CI 0.5?0.8} crises. Suicide decedents without known MH problems had significantly greater odds of leaving a suicide note 1.2, 95% CI while decedents with known NIH problems more often had a history ofsuicidal ideation {40.3% vs. 23.0%, .01} and attempts [29.4% vs. 10.3%, pg DISCUSSION [9201}?00 wordsj The current study highlights state-specific trends in suicide among people a 10 years, over time by state and by bed. It also provides a snapshot of circumstances of suicide among people with and without known mental health E565: Erin's: it}; 13 areas; i gear Es" assessment i132" 'LisTf i 'tF?i?r'e'ti 36 '51} if oil: .1 rates increased significantly, overall and among both males and females, aligns with past studies reporting on rate increases over time and across demographic subgroUpsF'f-l To extend this information, novel state?specific trend analyses identified overall suicide rate increases in 44 states. Five remaining states and D.C. reported non-significant rate increases and Nevada, showed an overall decline in suicide rates. The geographic distribution of rate increases suggests that many regions already known to have the highest rates of suicide in the country leg. mountain west], also experienced some of the largest percent increases in ?f suicide rates. Conversely, regions that once seemed more protected from suicide leg. northeast, midwest] also epidemic that occurred over the same time period?- and the downturn in the economy in the period 2002?20099 I Fame studies also suggest a negative impact of social media, particularly among youth. 11 Asha?s paper Rev 3.2.13 Related to gender, female suicide rates increased in more states {43] than did male suicide rates This may Enrolment '5 this too pointed or hint at a narrowing of the suicide gender gap, as has been found related to suicide in other demographic strongg groupings by urbanization It also aligns with reports of large increase in middle aged female suicides between 19992014.? This may signal an increases in risk factors more common among females intimate partner violence) or may reflect a differential impact of common suicide risk factors substance . :3 Comment This is a I different way to present this info. What if do you think? abuse} by sex. More research on this topic is needed. ;[Comment '5 it "idiom"? Mental illness is an import risk factolr for suicide. Unlike other studies,12 the current study found that more than representative atthis point? Or no? half of people did not have a known mental health problem. Among people who did, just over half were i known to be in treatment at the time of their deaths, about one-third were reported to never have received any mental health treatment, and between about 304096 had a history of ideation andjor suicide attempts. This ,i points to the need for increased identification of, and improved responses to, people at risk, and highlights the 5' need for better access to care that is evidence-based collaborative}, affordable, and sustained in order to l: prevent suicide attempts, re-attempts, and suicide. While there is a tendency to focus primary attention to the risk fact of mental illness, risk factors for suicide are manlfoldl?tratifying analyses by suicides among people with and without known mental health problems allowed 31351.6? 156 htri sass; i115 Ellie}? GEg'?EE'Fri?'riir differences in the prevalence of circumstances, for example people without mental health problems were more likely to have criminal-legal problems, evictionfloss of home, and recent crises. However, some circumstances such as intimate partner problems, arguments/conflicts, recent crises, and physical health problems are common across groups. Whether the two groups are viewed as more similar or more unique, results are clear that multiple challenging life circumstances impacted both. Both state trends and circumstance findings point to the same solutionche need for a comprehensive public health approach to suicide, inclusive of multiple and multi-Ievel evidence?based strategies and approaches that: i} prevent suicide risk before it occurs, by strengthening economic supports, strengthening access to, and coverage of, care that is safe and effective; creating protective environments leg. reducing access to lethal means among people at risk); promoting connectedness, and teaching coping and problem?solving skills; ii] Ii; I identify and support people already at risk, e.g. through gatekeeper training, crisis intervention, referral to i effective treatments, and prevention of re-attempts; and lessen harms after a suicide supporting 5 survivors and promoting responsible media reporting to avoid suicide contagionFEs of yet, no state has able to implement such a comprehensive approach. Limitations The findings provided of this report are subject to at least two limitations. Data for state level analyses extend to 2016, the date for which the most recent mortality data is available, and clata from come from 2015 and cover only 2? state% Abstractors of data are limited to data included in the investigative reports they receive. For captured diredtly-from medical'rrecb'r?ds'but from coroneri'medical examiner reports and the decedent?s family members and friends. Therefore, the completeness and accuracy of this information are limited by the knowledge of the informant. This may explain Need ref here 1? Need ref '12 13 Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: a technical package of policies, programs and practices. Atlanta, US Department of Health and Human Services, 201?. Rev 3.2.13 some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies.?1 Conclusions Suicide is a growing public health problem with states experiencing large increases in rates between 1999 and 2016. Mental illness is an important riskfactor for suicide and it is one of many factors. Comprehensive suicide prevention strategies are needed in states that encompass risk factors at multiple levels and that work to prevent suicide risk before it occurs, to identify and successfully support people atvrisk, and that lessens future harms. Corresponding author: Deborah M. Stone, Division of Violence Prevention, National Center for Injury Prevention and Control, CDC. Telephone: 77?0-488-3942; E-mail: zaf9@cdc.gov References Tables and figures 1" Need ref here Rev 3.6.13 Draft 1 Structured abstract {2441,1250 words) Background: Suicide rates have been asking?rising in the United States Examining state?level trends in, and contributing circumstances to, suicide, can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates, everall?and?bytsea?andstates by state and saga, among people r' aged 210 years, were eal valuate using data from the National Vital Statistics 5ystem?Elia-nge-s-in-Fate-sng? and state rankings were assessed across six consecutive 2005-2007, 2008-2010, 2011-2013, 2014-2016}. Data from the 2015 National Violent Death Reporting System, across 2? states, were analyzed to compare circumstances between suicide decedents with and without known mental health problems. Results: I. . significant upward rate trends were identified for 44 states. for the U5. overall and for 30 statesihdi?ddaillinu Comment [snlc?h LThis is the key strati?cation in my vie L. Comment lsnk?l: The word "assessed? would also be fine here. Comment [snkl?ah Probably shouldn?t mix the inferential findings with the descriptive findings in the same sentence. Also, we might not be able to afford the word space in the ,Abstract to talk about individual states. Comment And for 29 of these 30 states, the modeled rate trends were also statistically signi?cant. So these two sentences together, while keeping the inferential and descriptive findings sm?irlca' rates increased hr at least 25% we the Mr r' mental health problems experienced a range of contributing circumstances to their suicides, including recent crises and problems related to substance use, relationships, job/?nancial issues, criminal-legal matters and recent. Conclusions: Suicide rates have risen significantly in the 0.5. and across most states from 1999?2016. No single factor alone contributes to suicide. Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population~based public health approach inclusive of evidence?based strategies across multiple levels (individual, familyr'relationship, community, societal], focused on preventing risk before it starts, identifying and supporting people already at risk, preventing-reattempts, and caring for survivors after a suicide. INTRODUCTION BACKGROUND AND PURPOSE {210f250 words} TOTAL In 2016, nearly 45,000 suicides occurred in the United States While overall rates have been declining globally,2 rates of suicide in the U.S. have increased between 1999 and 2016, among males and females, across racialfethnic groups, and across urbanization levels.? Emergency department visits for nonfatal self-harm injuries increased by more than 40% between 2001 and 2015.5 In 2015, suicides and self-harm injuries cost the nation more than @369 one factor, rather, the risks are often numerous and occur at multiple levels?individual, familyfrelationship, community. and societal.? Despite this, [oi-evention primarily centers on mental disorder}. Other factors associated with suicide include social isolation, economic downturns, access to lethal means {93- skills, loss of a friend or family member to suicide, a prior suicide attempt, and unsafe media portrayals, among others.6 While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states struggle to make this a reality.? To better assist states, this study analyzes trends in State suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. (2591250 words} ,summary. r[Comment Need to update {Comment Need reference here. Comment Do we want to say that we can tie this in?? separate, provide a scientifically accurate opioids as an example here instead so Jk. 3 i Rev 3.6.18 Draft 1 Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Ciossificotion of Diseases 10?] Revision underlying-cause-of death codes X60-X84, Y87.0, Age-specific population estimates were obtained from US. Census Bureau National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2016. Rate estimates were age?adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from 2015 from the 27' states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics, including precipitating circumstances, of deaths by suicide among decedents with and without known current mental health problems (MHP). Mental health problems are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured in separate variables). compiles data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and racelethnicity. RESULTS (696.1600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rates trends were found to be significantly increasing for 44 states, as well as for the U.S. overall (Table 1). By sex, rate trends were found to be significantly increasing in 34 states for males and in 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known mental health problems (MHP) were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval and significantly greater odds of being raciallethnic minorities (odds ratio range: CI range [1.04.3] - They also had significantly greater odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the U.S. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.8% vs. most frequently by over-the-counterfotherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without known MHP had known precipitating circumstances. Decedents without known MHP had significantly greater odds of any type of relationship problem 1.3, 95% CI 1.214), and specifically of intimate partner problems 1.4, 95% 2 Rev 3.6.13 Draft 1 CI arguments or conflicts 1.4, 95% CI and having recently perpetrator interpersonal Comment We also have some violence iaDR 2D, 95% CI Two-thirds of decedents with known MHP had a history of MH or . additional data that is forthcoming on the types of recent crises, FYI so that may be substance abuse treatment [62.2% i, and were more likely to have any substance abuse problems (31.6% vs. added as well. 1 25%, 5 .01), i Suicide decedents without known MHP had significantly greater odds of other life stressors, such as a criminal legal problems 1.2, 95% Cl or evictionfloss of home laOR 1.4, 95% CI They had _l significantly lower odds of recent release from any institution, but when a release was indicated, they were i significantly more likely to be released from a correctional facility 4.5. 95% CI or hospital iaDP. 1.3, 95% CI Among decedents with known MHP who were recently released from an institution, i 42.8% release was from a institution. Those with known MHP also more frequently had job andfor i i financial problems [16.8% vs. 15.6%; 5 Decedents without known MHP had significantly greater odds of a crisis laClP. 1.4, 95% CI When the type of crisis was known, it was most frequently a problem related to an intimate partner physical health criminal legal issues a family relationship or a job Over one-fourth of decedents with a known MH problem also had recent or impending crises, most frequently related to problems with an intimate partner physical health or a family relationship Decedents without known MHP had significantly greater odds of criminal legal problems iaDFt 1.6, 95% CI and significantly lowered odds ofjob-related {aOFt DJ, 95% CI 0.5-0.8} crises. Suicide decedents without known MHP had significantly greater odds of leaving a suicide note iaDR 1.2, 95% CI while decedents with known MHP more often had a history of suicidal ideation {40.3% vs. 23.0%, .01] and attempts [29.4% vs. 10.3%, 5.01}. Conclusions and words} During the time period 1999?2016, age-adjusted suicide rates among people a it] years increased 25.4% overall Forty-four states saw significant rate increases and one {Nevada} state saw a significant decline. Suicide rates increased by more than 25% in 30 states and upwards of 50% in some. Among females, rates increased in 43 states and rates among males increased in 34. This signal of increasing vulnerability of females towards suicide aligns with recent reports that identified a 53% increase in middle-aged female suicide rates between 1999-2014 and an annual increase of 18.3% per year in emergency department visits for self-inflicted injuries among young females, aged of 10 and 14, in the period 2009-2015341? These increases may hint at a narrowing of the suicide gender gap, historically weighted towards males by a ratio of 4- 5:1.11 More rEsearch into this troubling trend is needed. One important factor associated with suicide is mental health problems. Nearly half of people in this study had a known mental health problem. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two?thirds of people with MHP had a history of mental health andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed. This includes the need for broader implementation of affordable and evidence-based treatments, such as doctor-patient collaborative care models and cognitive?behavioral therapy.? Additionally, greater ?tness to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. ?1 Ftev 3.6.13 Draft 1 While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MPH. This group suffered more life stressors, especially related to relationships leg. intimate partner problems, arguments or conflicts, recent perpetration of intimate partner violence}, but also related to other life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises [often related to the abovementioned factors). People with known MHP also experienced life and other stressors apart from their MPH. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone. These strategies may include: strengthening economic supports leg. housing stabilization policies household financial support}, teaching coping and problem-solving skills and other pro-social norms, especially early in life to manage everyday stressors and to prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments leg, reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help-seeking and positive social norms], and supporting people after a suicide has taken place to prevent survivors? risk and to assure safe reporting by the media in order to prevent suicide contagion}2 The findings provided of this report are subject to at least three limitations. In four states, Maryland Utah (UT). Massachusetts and Rhode Island state rankings might have been impacted by large proportions of deaths of undetermined intent or by decreased percentages of undetermined deaths over time (UT. MA, le. Second, data for state level analyses extend from 1999 to 2016, however, data on circumstances of suicide come from a single year [2015} and encompass only dataare limited to data included in the intrestigative reports they receive. For example, medical and mental health information are not captured directly from medical records but from coronerj?medical examiner reports and the decedent?s family members and friends. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. {this may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies. 13] Suicide is a growing public health problem and mental illness Is an important risk factor for suicide, but is just one of many associated factors. Resources such as CDC's Preventing Suicide: a Technical Pockoge of Policies, Programs. and Practicesnand the National Violent Death Reporting System can help states and communities prioritize comprehensive suicide prevention. References 1 Wisqars fatal injury reports 2 World report 3 NCHS Data brief 2014 4 Kegler et al 201? 5 Nonfatal injury reports 5 National Strategy for Suicide Prevention 7 Davidson, L, Potter, L., and Floss, {19991 Surgeon General?s Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 3 9 10 Melissa's paper 11 Need ref 12 Technical package Comment Is it nationally regreSentative at this point? Dr no? [Comment Keep thisRev 3.6.13 Draft 1 Acknowledgments Corn ms and Policyr Conflict of Interest None Corresponding author Stone Tables and Figures Rev 3.6.13 Draft ured b5 1! words) I[immanentjSIl?. Need to update Background: Suicide rates have been risking in the United States Examining state-IEvel trends in, and Comment [5111]: Need reference here. contributing circumstances to, suicide, can inform comprehensive state suicide prevention planning. Gamma?, Do we want to 53,, opioids as an example here instead so Methods: Trends in age-adjusted suicide rates, overall and by sex and state, among people aged 210 years, . that we can tiethis were calculated using data from the National Vital Statistics System. Changes in rates and state rankings were assessed across six consecutive three-year periods [1999-2001, 2002-2004, 2008-2010, 2011-2013, ll 20142015}. Data from the 2015 National ?v'iolent Death Reporting System, across 2? states, were analyzed to E: compare circumstances between suicide decedents with and without knoWn mental health problems. all Results: Average annual percentage change in suicide rates increased significantly in the 0.5. overall and in 44 states, with relative increases ranging from 5.9% in Delaware to 516% in North Dakota. People with M, and without known mental health problems experienced a range of contributing circumstances to Elli their suicides, including recent crises and problems related to substance use, relationships, jobffinancial issues, l3 criminal-legal matters and recent. if Conclusions: Suicide rates have risen significantly in the 0.5. and across most states from 1999?2016. No single ill factor alone contributes to suicide. I'll?; Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population-based all: public health approach inclusive of evidence-based strategies across multiple levels (individual, familylrelationship, community, societal], focused on preventing risk before it starts, identifying and supporting people already at risk, preventing-reattempts, and caring for survivors after a suicide. INTRODUCTION BACKGROUND AND PURPOSE {210f250 words) TOTAL coumasoo?soo l" lg: In 2016, nearly 45,000 suicides l13.4i100,000l occurred in the United States While overall rates have been declining globally,2 rates of suicide in the 0.5. have increased between 1999 and 2016, among males and females, across racial/ethnic groups, and across urbanization levels.M Emergency department visits for nonfatal til self?harm injuries increased by more than 40% between 2001 and 2015.5 In 2015, suicides and self-harm injuries we the nation more than [$69 one factor, rather, the risks are often numerous and occur at multiple levels-individual, familyfrelationship, 5: community, and societal.? Despite this, prevention primarily centers on mental disorder}. Other factors associated with suicide include social isolation, economic downturns, access to lethal i means skills, loss of a friend or family member to suicide, a prior suicide attempt, and unsafe media portrayals, among others}? While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states struggle to make this a reality.? To better assist states, this study analyzes trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METi-Ioos (2591250 words} Suicide rate estimates and trend analyses exclude data for persons :10 years old. Age-specific suicide counts Were tabulated based on National Statistics System coded death certificate records {international Classification of Diseases 10?" Revision underlying-cause?of death codes X60-X34, Age-specific population Rev 3.6.13 Draft 1 estimates were obtained from US. Census Bureau I National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics, including precipitating circumstances, of deaths by suicide among decedents with and without known current mental health problems (M Mental health problems are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured in separate va riables). compiles data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology}, and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and racefethnicity. RESULTS (696,!600 words) The most recent overall suicide rates (representing 2014?2016) ranged from 6.9 (District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to +57.6% (North Dakota), with percentage increases of at least 25% observed in over half of all states as well as nationally. Modeled suicide rates trends were found to be significantly increasing for 44 states, as well as for the U5. overall (Table 1). By sex, rate trends were found to be significantly increasing in 34 states for males and in 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known mental health problems (MHP) were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval [Cll and significantly greater odds of being racialfethnic minorities (odds ratio range: 1.0~2.1; 95% CI range They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the US. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.8% vs. most frequently by over-the-counterfotherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without known MHP had known precipitating circumstances. Decedents without known MHP had significantly greater odds of any type of relationship problem 1.3, 95% CI and specifically of intimate partner problems 1.4, 95% CI arguments or conflicts 1.4, 95% CI and having recently perpetrator interpersonal violence 2.0, 95% CI Two-thirds of decedents with known MHP had a history of MH or 2 Rev 3.6.18 Draft 1 substance abuse treatment (67.2% and were more likely to have any substance abuse problems (31.6% vs. 25%. 5 ?ll. Suicide decedents without known MHP had significantly greater odds of other life stressors, such as a criminal legal problems 1.7, 95% CI or eviction/Toss of home 1.4, 95% They had significantly lower odds of recent release from any institution, but when a release was indicated, they were significantly more likely to be released from a correctional facility 4.5, 95% CI or hospital 1.3, 95% CI Among decedents with known MHP who were recently released from an institution, 42.8% release was from a institution. Those with known MHP also more frequently had job andfor financial problems (16.8% vs. 15.6%; 5 .05). Decedents without known MHP had significantly greater odds of a recentfimpending crisis 1.4, 95% CI When the type of crisis was known, it was most frequently a problem related to an intimate partner physical health criminal legal issues a family relationship or a job Over one-fourth of decedents with a known MH problem also had recent or impending crises, most frequently related to problems with an intimate partner physical health or a family relationship Decedents without known MHP had significantly greater odds of criminal legal problems 1.6, 95% CI and significantly lowered odds ofjob-related 0.7, 95% CI 0.508) crises. Suicide decedents without known MHP had significantly greater odds of leaving a suicide note 1.2, 95% CI while decedents with known MHP more often had a history of suicidal ideation (40.8% vs. 23.0%, ,o 5 .01) and attempts (29.4% vs. 10.3%. .0 .01) Conclusions and Comments (7351700 words) During the time period 1999-2016, age?adjusted suicide rates among people a 10 years increased 25.4% overall. Forty~four states saw significant rate increases and one (Nevada) state saw a significant decline. Suicide rates increased by more than 25% in 30 states and upwards of 50% in some. Among females, rates increased in 43 states and rates among males increased in 34. This signal of increasing vulnerability of females towards suicide aligns with recent reports that identified a 63% increase in middle?aged female suicide rates between 1999-2014 and an annual increase of 18.8% per year in emergency department visits for self-inflicted injuries among young females, aged of 10 and 14, in the period 2009-2015110 These increases may hint at a narrowing of the suicide gender gap, historically weighted towards males by a ratio of 4? 5:1.11 More research into this troubling trend is needed. One important factor associated with suicide is mental health problems. Nearly half of people in this study had a known mental health problem. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two?thirds of people with MHP had a history of mental health andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed. This includes the need for broader implementation of affordable and evidence-based treatments, such as doctor?patient collaborative care models and cognitive?behavioral thera py.12 Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. 12 While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MPH. This group suffered more life stressors, especially related to relationships intimate partner problems, arguments or conflicts, recent perpetration of intimate partner violence), but also related to other 3 Rev 3.6.13 Draft 1 life stressors such as criminalvlegal matters, eviction/loss of home, and recent or impending crises [often related to the abovementioned factors]. People with known MHP also experienced life and other stressors apart from their MPH. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone. These strategies may include: strengthening economic supports leg. housing stabilization policies, household financial support}, teaching coping and problemvsolving skills and other pro-social norms, especially early in life to manage everyday stre55ors and to ore-vent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help-seeking and positive social norms], and supporting people after a suicide has taken place to prevent survivors' risk and to assure safe reporting by the media in order to prevent suicide contagion.? The findings provided of this report are subject to at least three limitations. In four states, Maryland Utah Massachusetts and Rhode Island (RI), state rankings might have been impacted by large proportions of deaths of undetermined intent or by decreased percentages of undetermined deaths over time (UT, MA, Rll. Second, data for state level analyses extend from 1999 to 2015, however, data on circumstances of suicide come from a single year [2015] and encompass only limited to data included in the investigative reports they receive. For example, medical and mental health information are not captured directly from medical records but from coroner/medical examiner reports and the decedent's family members and friends. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies. 13 Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but ls just one of many associated factors. Resources such as Preventing Suicide: 0 Technical Package of Poiicies, Programs, and Procti'ces?zand the National Violent Death Reporting System can help states and communities prioritize comprehensive suicide prevention. References BE Wisqars fatal inlury reports World report NCHS Data brief 2014 Kegler et al 2D17 WISCIARS Nonfatal injury reports National Strategy for Suicide Prevention Davidson, L., Potter, and Ross, V. {1999} Surgeon General's Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 3 9 10 Melissa?s paper 11 Need ref 12 Technical package Acknowledgments Con?ict of interest None Corresponding author Stone Tables and Figures [Comment Is it nationally representative at this pointrt Comment Should we say anything about the time periods we?re ,referring to here? Comment if we have another few words, might want to say that some common and some differing circumstances contribute to suicides of those with and without known mental health problems, since that's a big focus for us. I know we're tight on words. just a Lsuggestion if possible {Comment Could we shorten to "primary prevention"? 4 Comment IFKAU: i think a lot of people not familiar with this area think of survivors as people who themselves fl survived a suicide attempt. Can We clarify if with something like leg, friends and family of suicide decedents}, or i something more plain language? i: ll [Comment Need to update l[I'Somment Need reference here. ., Fiev 3.6.13 Draft 1 Structured abstract {2441,1250 words) Background: Suicide rates have been risking in the United States Examining state?loyal trends in, and contributing circumstances to, suicide, can inform comprehensive state suicide prevention planning. 5b?; Methods: Trends in age-adjusted suicide rates, overall and by sex and state, among people aged 3:10 years, were calculated using data from the National Vital Statistics System. Changes in rates and state rankings were assessed across six consecutive threeayear periods (1999-2001, 2002-2004, 2005-2001 2008-2010, 2011-2013, 2014-2016}. Data from the?2015 from the National Violent Death Reporting System, across 27 states, were 1 analyzed to compare precipitating circumstances between suicide decedents with and without known mental health problems. Results: Average annual percentage change in suicide rates increased significantiylin the US. overall and in 44 states, with relative increases ranging from 5.9% in Delaware to 51.5% in North Dakota. People with and without known mental health problems experienced a range of eonsFi-beei-ng?circumstances contributing to their suicides, including recent crises and problems related to substance use, relationships, job/financial isSues, and criminal?legal Conclusions: Suicide rates have risen significantly in the 0.5. and across most states from 1595-2016. [do single i Elf,- Comment Probably could cut these examples of diagnoses as long as factor alone contributES to ing those with Livili'ic-u'. mental Ilee?th problems. Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population?based public health approach inclusive of evidence-based strategies across multiple levels (individual. . family/relationship, community, societal), fowsed on breventing risk before it starts}, identifying and supporting f; u. people already at risk, preventing?reattempts, and caring for Eurvivari after a suicide. :le the reference gets into de?ning mental . lfr' ,Illness sufficiently. 1 I'll?! i Comment Do we want to say Hi opioids as an example here instead so 4! ,that we can tie this in?? I I Comment l?d vote to leave it broad, orjust consistent with the National INTRODUCTION LStrategy which we reference here. BACKGROUND AND PURPOSE {210(250 words} TDTAL 1300 In 2016, nearly 45,000 suicides {114,000,000} occurred in the United States While overall rates have been declining globally} rates of suicide in the US bane?increased between 1999 and 2016, among males and females, across racial/ethnic groups, and across urbanization levels?? Emergency department visits for nonfatal self?harm injuries increased lay?more than 40% between 2001 and 2015.5 In 2015, suicides and self?harm I injuries we the nation more than {$59 one factor; rather, the risks are often numerous and acour at multiple levels?individual. familylrelationship, community. and societal? Despite this, prevention primarily centers on mental illnesslliegu depression, bipolar means ubstanced, firearms) among people at risk, childhood adversity, lack of coping and problem?solving skills, loss of a friend or family member to suicide, a prior suicide attempt, and-Unsafe media portrayalsrafnong others" While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states struggle to make this a reality.? To better assist states, this study analyzes trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METHODS words} Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age?specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {international Classification of 1 Rev 3.6.13 Draft 1 Diseoses 1'0?1 Revision underlying-cause-of death codes X60-X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureau National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three? year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics, including precipitating circumstances, of deaths by suicide among decedents with and without known current mental health problems (M Mental health problems are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception ofalcohol and other substance dependence (captured eempilesaggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racey?ethnicity. RESULTS (696i600 words) The most recent overall suicide rates (representing 2014?2016} ranged from 6.9 (District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to +5.16% (North Dakota), with percentage increases of at least 25% observed in over half of all states as well as nationally. Modeled suicide rates trends sigr:.iitanlly for 44 states, as well as forthe U.S. overall (Table By sex, rate trends increasel significantly ima?eas-leg?in 34 states for males and in 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +16% for females. Suicide decedents with and without known mental health problems (MHP) were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval (C11 and significantly greater odds of being racialfethnic minorities (odds ratio range: 102.1; 95% Cl range They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio [303] 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% Cl Fifteen percent of those with known MHP and 20% without ever served in the U.5. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP {19.8% vs. most frequently by overethe-counteriotherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without mews?MHP had known precipitating circumstances. Decedents without known MHP had significantly greater odds of any type of relationship problem 1.3, 95% CI sad?specifically el?intimate partner problems 1.4, 95% CI argumentsg?er?conflicts 1.4, 95% Cl and la-awng recently perpetratejgger 2 Rev 3.6.13 Draft 1 {Comment IvidSI: Could probably cut this] i if we need to for space interpersonal violence 21], 95% CI 1.324}. Two~thirds of decedents with known MHP had a history of MH or substance abuse treatment {62.2% J, and were more likely to have any substance abuse problems [31.6% Comment [vidS]: Maybe we should be more precise Suicide decedents without known lleP had significantly greater odds of other life stressors, such as a criminal legal problems 1.2, 95% CI or evictionfloss of home 1.4, 95% CI They had significantly lower odds of recent release from any institution, but when a release was indicated, they were significantly more likely to be released from a correctional facility 4.5, 95% CI or hospital [ani 1.3, 95% CI Among decedents with known MHP who were recently released from an institution, 42.8% from institutions. Those with MHP also more frequently had job vs. 25%, .01), i i I andg'or financial problems [16.8% vs. 15.6%; .05). Decedents without known MHP had signi?cantly greater odds ofa recentfimpending crisis laDR 1.4. 95% CI When the type of crisis was known, it was most frequently senseless-related to an intimate partner physical health criminal legal issues a family relationship; or a job Over 3' one-fourth of decedents with a known also had recent or impending crises, most frequently i il I. ,ll :5 ii related to arehleeas?weh-an intimate partner physical health or afamlly relationships [Decedents without known MHP had signi?cantly greater odds of criminal legal p: obiems tilt-aglaOR 1.6, 95% CI and significantly lowered odds ofjob?related . "cat" laDR I117, 95% CI 0.5-0.3lc-rec-a CI 111.2}, while decedents with known more often had a history of suicidal ideation {40.8% vs. 23.0%, 5 .01] and attempts [29.4% vs. 10.3%, ,o 5 Conclusions and Comments words] During the lime 1999-2015, age?adjusted suicide rates among people a 10 years increased 25.4% overall. Forty?four states saw significant rate increases and one {Nevada} state saw a significant decline. Suicide rates increased by more than 25% in 30 states and upwards of 50%[n somt?. Among females, rates increased in 43 states and rates among males vulnerability of females towards suicide aligns with recent reports that identified a 53% increase in middle-aged female suicide rates between 1999-2014 and an annual increase of 13.3% per year in emergency department visits for self-inflicted injuries among young females, aged of 10 and 14, in the period These increases may hint at a narrowing of the suicide gender gap, historically weighted towards males by a ratio of 4- i I i 5:1}1 More research into this troubling trend is needed. One important factor associated with suicide is mental health problems. Nearly half of people in this study had a known mental health problem. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two?thirds of people with MHP had a history of mental health andy'or substance abuse treatment and over half Were currently in treatment at the time oftheir deaths, much more support for this vulnerable population is needed. This includes the need for broader implementation of affordable and evidence?based treatments, such as doctor?patient collaborative care models and cognitive-behavioral therapy.12 Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. ?2 While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MPHE. This group suffered more life stressors, especially related to relationships intimate partner I a Fiev 3.6.13 Draft 1 Erroblems, arguments or conflicts, recent perpetration of intimate partner violence}, but also related to other life stressors such as criminalvlegal matters, evictionfloss of home, and recent or impending crises [often related to the abovementioned also experienced life and other stressors apart from their This group was more likely to experience lob safaris;tastier;reassessment; intimate partner problems l24.1%l, physical health problems and recent or impending crises These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone. These strategies may include: strengthening economic supports leg. housing stabilization policies, household financial support}, teaching coping and problem?solving skills and other pro?social norms, especially early in life to manage everyday stressors and to prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments leg, reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help-seeking and positive social norms], and supporting people after a suicide has taken place to prevent survivors? risk and to assure safe reporting by the media in order to prevent suicide contagion.1L2 Thesr: findings at least three limitations. In four states, Maryland Utah Massachusetts and Rhode island state rankings might have been impacted by large proportions ofdeaths of undetermined intent (MD), [which often represent cases wheie a suicide determination worried ?at is be Second, is not yet nationally representative. Currently, 40 states, the District of Columbia, and Puerto but [lit?3 mus! retelijvailalile data year Includes 2? state; as llit' systerri Ir-i terdata-ieesiate isles-lea FiFi some from a single year and encompass only?istatea Third, abstractors of data are limited to data included in theinvestigative reports-Heey-reteive. For example, medical and mental health information are not captured directly from medical records but from leg, family, friends} via coroner/medical examiner g. 'i I .- r. reports and the decedent?s family members and friends. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies.[13] a I I. 7: Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many associated factors. Resources such as Preventing Suicide: Technicoi Package of Policies, Programs, and Procticeslzand the National Violent Death Reporting System can help states and communities prioritize comprehensive suicide prevention. I 1 1 I I References BE 1 Wisqars fatal injury reports 2 World report 3 NCHS Data brief 2014 4 Kegler et al 201? 5 WISQARS Nonfatal injury reports 6 National Strategy for Suicide Prevention 7 Davidson, L., Potter, L, and Ross, v. [1999} Surgeon General's Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 8 ICU-10 9 DEM-5 1IJ Melissa?s paper 11 Need ref g, 1 ?ilrepresentative at this point? 0r no? I Comment IvidSI: Wondering if instead of recapping results, we could comment on the profound and not often discussed impact that life stressors, especially those that rise to the level of a crisis, can have. The research that shows the time amount of deliberation [which is often very low) may also be helpful to cite here, as these types of stressorsicrises can represent short?term problems (vs. longer term Lproblems such as chronic mental illness) 1 Comment vid5l= Similarly, wondering if we could condense this instead of reiterating some of these results and instead comment on the concept of "functional impairment,? as a reason why people with mental health problems might have more jobffinancial problems le.g., they often have more inconsistent I 1 . . I . Comment Ivid?l: I know we don?t want to really be adding words but I thought this is one thing that may not be obvious to everyone {why deaths of undetermined intent would affect suicide See what you think, and if we can't get the text in, maybe we could ?sneak a footnote in JL. Comment Is it nationally Comment [vidSlt Some suggested rephrasing here, because i don?t want to trigger questions about why we would only use one data year of data. We have good reasons more states, and for the state level analyses we were looking at trends and therefore needed more years. wondering if WE I I . Comment IvidSI: Just to give our data enough credit, though, I am wondering if the 9096 estimates could also be driven up by small or very selective samples. Was going to look into this but not sure what the reference is. Little wary of making it gsound like theirs is the gold standard. Rev 3.6.13 Draft 1 12 Technical package Acknowledgments Conflict of Interest None Corresoonding author Stone Tables and Figures Fiev 3.6.13 Draft 1 Structured abstract {241,(250 words) Background: Suicide rates have been risking in the United States Examining state?level trends in, and contributing circumstances to, suicide, can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates, overall and by sex and state]: were awesomeness rankings were assessed across six consecutive periods {1999-2001, 2002?2004, 2005?2001, 2003? 2010, 2011-2013, 2014-2016}. Data from the2015 from the National Violent Death Reporting System, across 2? states, were analyzed to compare precipitating circumstances between suicide decedents with and without {Comment Scott?s comment: I Comment Scott's comment: The word assessed would also be fine here. Comment lsnk?l: And for 29 of these 30 states, the modeled rate trends were also inferential and descriptive findings his is the keystratification in my view. all. tatistically significant. So these two entences together, while keeping the known mental health problems. Ir separate, provide a scientifically accurate Results: Statistically significant upward rate trends were identified for gistares [for the oye_r_a_il_sntl _fo_r_3__0 :summary. I a slates individually, empirical rates increased by at least 25%- over the study period] Enter-age annual percentage 5' Comment Wm?: ShC'Uld WE 53? c?leiar?igee'r .E, "r .1 i frank 1-H Hat 1 5 ar? 4% anything aboutthe time periods we?re .Il W- I I. referrin to here? ranging .st in-Qelawate With and Without known mental 1 health problems experienced a range of eentnbutingcircumstances contributing to their suicides, including Comment ISDH: Scott?scomments recent crises and problems related to substance use, relationships, job/financial issues, and_criminaI-legal EroPahlv Show? m'?thfg mferej'nhal, tt nd? findings with the descriptive findings in ma ers-a rece the same sentence. Also, we might not Conclusions: Suicide rates have risen significantly in the 0.5. and across most states from 1999?2015190 single factor alone contributes to suicide.]_ Differing circumstances contribute to suicides among I without mental health problems. Comment IFKAH: If we have another few words, might want to say that some Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population?based f?mm?? and some differing ?l Al I public health approach inclusive of evidence-based strategies across multiple levels (individual, family/relationship, community, societal], focused on breventing risk before it start? identifying and supporting people already at risk, preventing?reattempts, and caring forEurvivor? after a suicide. . -. ?li INTRODUCTION BACKGROUND AND PU RPDSE {2101'250 wards] TOTAL In 2015, nearly 45,000 suicides {lid/100,000) occurred in the United States While overall rates have been declining globally,2 rates of suicide in the US. hassle?increased between 1999 and 2016, among males and females, across racial/ethnic groups, and across urbanization lt?zevels?"1 Emergency department visits for nonfatal self?harm injuries increased bymore than 40% between 2001 and 2015.5 In 2015, suicides and self?harm . injuries see the eetiee mere the" E359 one factor; rather, the risks are often numerous and occur at multiple eve s--individual, .r Comment [vid5]: lCould we shorten to F{Comment Need to update g" ,rLComment [50(1: Need reference here. primary prevention"? JL. Comment ICAI: i like "primary prevention" ., Comment think a lot of people not familiar with this area think of survivors as people who themselves Comment Options: -Rosenman SJ. Preventing suicide: What will work and what will not. MJA. 1 a. community, and societal}? Despite this, prevention primarily centers on mental illness?ie.g., depression, bipolar disorder?. Other factors associated with suicide include social isolation, economic downturns, access to lethal means {egEubstanced firearms) among people at risk, childhood adversity, lack of coping and problem-solving (z skills, loss of a friend or family member to suicide, a prior suicide attempt, and unsafe media portrayals, among others? While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states struggle to make this a reality.? To better assist states, this study analyzes trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention 1 recommendations. METHODS words} IL 1 Comment FKAilz Probably could cut these examples of diagnoses as long as Lthe reference gets into de?ning mental "i - Comment Do we want to say opioids as an example here instead so that we can tie this in?? 1 Comment I?d vote to leave it broad, or just consistent with the National Strategy which we reference here. Rev 3.6.13 Draft 1 Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationoi Ciossificotion of Diseases 10*? Revision underlying-cause-of death codes KEG-X84, Y87.0, Age-specific population estimates were obtained from U.5. Census Bureau National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2015. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics, including precipitating circumstances, of deaths by suicide among decedents with and without known current mental health problems (MHP). Mental health problems are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured compiles-aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and race/ethnicity. RESULTS (6961'600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 5.9 {District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to +57.6% (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rates trends significantly for 44 states, as well as for the U. 5. overall (Table 1). By sex, rate trends increased significantly 34 states for males and In 43 states for females. Nationally, the model- estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known mental health problems (MH P) were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval (CI) and significantly greater odds of being racialiethnic minorities (odds ratio range: 1.0?2.1; 95% Cl range They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio - 2.9, 95% - of firearm suicide positive toxicology results for alcohol (aDFi= 1.2, 95% Cl- Fifteen percent of those with known MHP and 20% without ever served in the US. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.3% vs. most frequently by over?the-countery?otherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without known?MHP had known precipitating circumstances. Decedents without known MHP had significantly greater odds of any type of 2 Rev 3.6.13 Draft 1 relationship problem 1.3, 55% CI and-specifically lat-intimate partner problems laOR 1.4, 95% CI arguments??aeconflicts 1.4, 95% Cl and basing?recently perpetratedinges interpersonal violence laDR 2.0, 95% CI Two-thirds of decedents with known MHP had a history of MH or substance abuse treatment (52.2% i, and were more likely to have any substance abuse problems [31.6% vs. 25%, .01), Suicide decedents without known had significantly greater odds of other life stressors, such as a criminal legal problems 1.2, 95% CI or evictioni'loss of home iaDR 1.4, 95% CI They had significantly lower odds of recent release from any institution, but when a release was indicated, they were significantly more likely to be released from a correctional facility 4.5, 95% CI or hospital laOR 1.3, 95% Cl Among decedents with known lleP who were recently released from an institution, 42.8% release?wasm from institutions. Those with known MHP also more frequently had job andfor financial problems {16.8% vs. 15.6%; 5 .05). Decedents without known MHP had signi?cantly greater odds ofa recent/impending crisis laOR 1.4, 95% CI When the type of crisis was known, it was most frequently a?pseblem-reiated to an intimate partner physical health criminal legal issues a?family relationships or a job Over one-fourth of decedents with a known MH-preblemMHP also had recent or impending crises, most frequently related to ?aelems?with?an intimate partner physical health or a?family relationships ?lletedents without known MHP had signi?cantly greater odds of criminal legal emblemsrgiseilao? 1.5, 95% CI and significantly loWered odds ofjob-related crises laOR 95% CI Suicide decedents without known MHP had signi?cantly greater odds of leaving a suicide note laDR 1.2, 95% CI while decedents with knOWn MHP more often had a history of suicidal ideation (40.3% vs. 23.0%, 5111) and attempts [29.4% vs. 10.3%, 5.01). Conclusions and Comments words) Qaeieg?the?eme?pesied?om 1999-2016, age?adjusted suicide rates among people a 10 years increased 25.4% overall. Forty-four states saw signi?cant rate increases and one {Nevada} state saw a signi?cant decline. Suicide Among females, rates increased in 43 states and rates among males increased in 34. This signal of increasing a vulnerability of females towards suicide aligns with recent reports that identified a 53% increase in middle?aged female suicide rates between 1999-2014 and an annual increase of 13.8% per year in emergency department visits for self-inflicted injuries among young females, aged of 10 and 14, in the period 2009?20153!? These increases may hint at a narrowing of the suicide gender gap, historically weighted towards malesiay?a?rat-re?ef??e- 11 More research into this troubling trend is needed. if we need to for space [Comment [vid5]: Maybe we should be more precise Comment ICAI: Maybe could leave out ratio to save space. [Comment IvidSI: Could probably cut this with suicide is mental health half of people in this study?hedo? known mental health problem. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two?thirds of peeple with leHP had a history of mental health andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed. This includes the need for broader implementation of affordable and evidence?based treatments, such as doctor-patient collaborative care models and cognitive-behavioral therapy.? Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. 13 i i i Rev 3.6.13 Draft 1 While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a Comment IvidSI: Wondering if instead known MPHE. This group suffered more life stressors, especially related to relationships tag. intimate partner I of recapping results, we could comment problems, arguments or conflicts, recent perpetration of intimate partner violence], but also related to other on the profound and ?011 Often life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises (often related .5 'mpad thatl'fe Stressc'rs' esf'ec'all" those to the abovementioned factorsll People with known MHP also experienced life and other stressors apart from that to the law? Of 3 ms?? can have" The research that shows the tIme amount their MPHP. This group was more likely to experience Eob and/or ?nancial problems. Also common were of deliberation {which is often VEW low) intimate partner problems physical health problems and recent or impending crises may ago be helpful to cite here, as these . . types of stressorsicrises can represent These results palm to the need for comprehensIve smcIde prevention that goes beyond a focus on MH an short?term problems lV5~ longer term treatment alone. These strategies may include: strengthening economic supports leg. housing stabilization ?a kproblems such as chronic mental illnesspollcles, household fInancIal support}, teachIng copmg and problem-solvmg skIlIls and other pro-socIal norms, . Comment IridSI: Similarly, wondering if especIally early in Me to manage everyday stressors and to prevent future relatIonshIp problems; and promotIng we could condense this instead of social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and reiterating some ofthese results and social support, as needed. Other strategies indicated by these results include creating protective environments instead comment on the William 0f "functional impairment,? as a reason why people with mental health problems might have more jobffinancial problems leg, they often have more inconsistent I . . I . Tl'lE? findings at least three limitations. In four states, Maryland I . . Comment IridSI: I know we don?t want (MD), Utah Massachusetts and Rhode island state rankings might have been Impacted by large I to really be adding words but I thought proportions of deaths of undetermined intent (MD), l?hich often represent cases where a suicide determination this is one thing that may not be obvious le.g., reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help-seeking and positive social norms), and supporting people after a suicide has taken place to prevent survivors' risk and to assure safe reporting by the media in order to prevent suicide contagion.12 ?5 was judged not to be conclusive lor by decreased percentages of undetermined deaths over time (UT, MA, Hi]. to everyone [why deaths or Second,- ityyogs Is_ not _yet nationally repr esemggiye Current tly, 40 statesthe District ofgpjombia and P_uer_to undetermined intent 3m?? SUlClde See what you think, and if we can't get the text in, maybe we could Rico participate in but the most recent available data year includes states as others mined the system ysneak a footnote in Third, abstractors of NUDRS data are limited to data [50ommcn sit nationa Included InIthe Investlgatlve reports-tome: For example, medical and mental health information are not ?x [representative atthis point? 0r no? I captured dIrectly from medlcal records but from Informants leg, famIly, friends] we coroner/medlcal exammer ?nr Therefore, knowledge of the informant impacts comm?? lynd5l: Some suggemd . . . . . rephrasmg here, because i don?t want to completeness and accuracy of the Informatlon rcported. This may explain some of the discrepancy betWeen the . . . . . I I trigger questlons about why we would prevalence of mental health dIsorders reported here and the 90% statistic frequently cited In other my?, use one data year of data. studies.[13] We have good reasons more states, and . . for the state level analyses we were SuIcIde Is a growmg publIc health problem and mental Illness Is an Important rIsk factor for but Is Just looking at trends and therefore needed one of many associated factors. Resources such as Preventing Suicide: a Technical Package of Policies, ?3 more years. However, wondering if we Programs, and Procticeslzand the National 'v?iolent Death Reporting System can help states and communities prioritize comprehensive suicide prevention. Comment ?5t ?3 give our data enough credit, though, I am wondering if References BE the 90% estimates could also be driven up by small or very selective samples. Was 1 Wisqars fatal injury reports going to look into this but not sure what 2 World report the reference is. Little wary of making it 3 NCHS Data brief 2014 Lsound like theirs is the gold standard. 4 Kegleret al 201? WISDARS Nonfatal injury reports 6 National Strategy for Suicide Prevention Davidson, L., Potter, and Ross, V. {19991 Surgeon General's Call to Action to Prevent Suicide. Public Health Service Rockyille, MD. 3 Rev 3.6.13 Draft 1 9 10 Melissa?s paper 11 Need ref 12 Technical package Acknowledgments Conflict of Interest None Corresponding author Stone Tables and Figures Options for ref 11 - :Hawton l1, var: Heeringen Ii. Suicide. Lancet. 2009.373: 13.72491. 3.2.13 -. Canetto SS, The Gender Paradox in Smcide 8L Life?Threatening Behavior_ 1998,- 28:1-23. From Katie: I went through the latest draft walex?s comments, comparing the results and discussion. Most lined up, although there were a few places where I noticed some mismatch is from the draft, bold are my comments]: ?From 1999-2016, age-adjusted suicide rates among people 2 10 years increased 25.4% overall.? We give the national (overall) number only in the abstract, not in the results. ?Suicide rates increased by more than 25% in 30 states and upwards of 50% in some?. We only mention one [North Dakota) that is over 50% in the results {rather than "some"l. "While two?thirds of people with MHP had a history of mental health and/or substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed.? The highlighted point is only in the table right now- could add to results text. "While are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MHP. This group suffered more life stressors, especially related to relationships intimate partner problems, arguments or conflicts, recent perpetration of intimate partner Didn?t notice this the first time around, but the highlighted portion should read ?recent perpetration of interpersonal violence" that variable is not speci?c to IPV. ?People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead). Rev 3.6.13 Draft 1 Stru?ured abstract {241]250 words) {[Cornment Scott?scommentThis is the ice stratification 'n view. Background: Suicide rates have been risking in the United States Examrning state?level trends in, and I contributing circumstances to, suicide; can inform comprehensive state suicide prevention planning. v: Comment ISDU: Scott's comment: The ,i ,word assessed would also be fine here. Methods: Trends in age-adjusted soicide rates, overallaaneigay state and sexand?stateLa Comment [sol-16!: years, were data from the National Vital Statistics System. Changes in rates and state And for 29 of these 30 states, the rankings were assessed across six consecutive three?year periods (1999-2001, 2002?2004, 2005-2002, 2003.? modeled rate trends were also 2010, 2011-2013, 2014-2016}. Data from the2015 from the National Violent Death Reporting System, across 23" statistically significant. 5? these two . . . . . . . . sentences together, while keeping the states, were analyzed to compare precipitating circumstances between soloide decedents With and Without . l' inferential and descriptive findings known mental health problems. If separate, provide a scientifically accurate Results: Statistically significant upward rate trends were identified for 44 states. [for the U.5. overall and for 30 :summary. .1 I a states individually, empirical rates increased by at least 25% over the study period] [averageennoei-pereentage Comment IFK?lli Should WE 53? - - 1" anything aboutthe time periods we?re Lreferring to here? Comment Scott?s comment: recent crises and problems related to substance use, relationships, jobfi?inancial issues, a_n_c_l_criminaI-lega EroPahlv Show? m'?thfe mferej'nha! 2' Comment If we have another Conclusions: Suicide rates have risen significantly in the LLB. and across most states from 1999?2015190 single few might Will? to Sail that Some . . . . . factor alone contributes to - Wtirizzorristaiices urge differ t: Comment IvidSI: Could we shorten to If L"primary prevention"? Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population?based If?, Cnmmem I like ?primaw public health approach inclusive of evidence-based strategies across multiple levels (individual, revention? familyfrelationship, community, societal], fooused on [:ireventing risk before it startg identifying and supporting j" Comment i think a lot of people people already at risk, preventing?reattempts, and caring forsugoortme Eur-Wvorslafter a-suicide tlecedents: not familiar with this are; think of surviving loved Comment IHKU: loose a solution to Katie's INTRODUWDN question here. BACKGROUND mo PURPOSE {21oizso words) TOTAL coum=1soor1aoo ?Bed Update 3 1 Should we add In 2016, nearly 45,000 smcides {13.4f100,000l occurred in the United States While overall rates have "treating? before mental ?mess? 1 been declining globally,2 rates of suicide in the U.S. have?increased between 1999 and 2010, among males and . . . . 34 . . i" ,{Comment Need reference here. females, across racialfethnic groups, and across urbanization levels. . Emergency department vi5its for nonfatal ,r ,r ,a 1 self?harm injuries increased eymore than 40% between 2001 and 2015.1 In 2015, suicides and self?harm Comment lCAl= 09??"55 injuries cost the nation more than E369 billionln direct medical and work loss costs.l Suicide is rarely caused by i it,? ?E?asenman 5i Prevent'm I [I?ll ?Ellandl?lhatlli" an: lulu one factor; rather, the risks are often numerous and occur at multiple eve s--individual, fa milylrelationship, ?i 2 Comment [FKililk Probably could cut community, and societal. Despite this, Earevention primarily centers on E'nental illnesiheg, depressmn, bipolar these examples of diagnoses as long as disorder} Other factors associated with suicide include social isolation, economic downturns, access to lethal WM means {egE-ubstancea firearms} among people at risk, childhood adversity, lack oinoor coping and problem- My" Comment Do we want to say solving skills, loss of a orig to suicide, a prior suicide attem pt, and unsafe media opioids as an example here instead so portrayals, among others.S While Although the Surgeon General called for a comprehensive public health 4: approach to suicide prevention in 1999, to date, most states struggle to make this a reality.l5 To better assist Comment l?d vote to leave it broad, or just consistent with the National states, this study analyzes trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. ivisri-ioos (ass/25o words} Rev 3.6.13 Draft 1 Suicide rate estimates and trend analyses exclude data for persons <10 yea rs old. Age-specific suicide counts were tabulated based on National 1lilital Statistics System coded death certi?cate records {international Classification of Diseases 10?? Revision underlying?cau5e?of death codes Age?specific population estimates were obtained from 0.5. Census Bureau [National Center for Health Statistics bridged-race population data releases. National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods yteesir on! 1999-2016. Rate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled ,l rate trends are reported in terms of average annual percentage changes i ll Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics[ including precipitating suicide among decedents with and without known current mental health problems Meata'riseai'ra pi are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence {captured data from three primary data sources: death certificates, coronerfrnedical examiner reports {including toxicology}, and law enforcement reports. Decedents with and without known MHP Were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% con?dence intervals controlling for age group, sex, and racex'ethnicity. RESULTS {696/600 words} The most recent overall suicide rates {representing 2014-2015} ranged from 69 {District of Columbia to 29.2 {Montana} per 100,000 persons per year. a four?fold difference {Table Across the entire study period, rates increased in all but one state {Nevada}, with increases ranging from +0.2 {Delaware} to +8.1 {Wyoming}. Percentage increases in rates ranged from +59% {Delaware} to {North Dakota}, with percentage increases of at least 25% observed in oVer half of all states as well as nationally. Modeled suicide rate trends Indicate increases for 4d states, as well as for the U.5. overall {Table By sex, modele?rate trends incseasedindicate increases in 34 states for males and in 43 states for ht females. Nationally, the modelvestimated AAPC for overall suicide rates was By sex, the national MPG was +11% for males and +15% for females. Suicide decedents with and without known were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male {95% confidence interval 2.2v2.5}, and significantly greater odds of being racialfethnic minorities {odds ratio range: 102.1,- 95% Cl range [1.013] - They also had significantly greater odds of perpetrating homicide-suicide {adjusted odds ratio 2.9, 95% CI of?rearm suicide 1.6, 95% El and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the 0.5. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without l'leP {19.8% vs. most freq uentiy by unclassi?ed drugs opioids antidepressants or benzodiazepines i this rimment lthinii you can delete Rev 3.6.13 Draft 1 All suicide decedents with known MHP and approximately 85% without had known precipitating circumstances. Decedents without known I'leP had significantly greater odds of any type of [specifically intimate partner or- I I i relationship problem (aUFl 1.3, 95% CI recently perpetrat CI arguments conflicts 1.4, 95% CI 1.315}, and interpersonal violence (aDFl 2.0, 95% CI 1.92.4). Twoathirds of decedents with known MHP had a history of MH or substance abuse treatment (67.2% )7 and were more likely to have any substance abuse problems (31.5% vs. 25%, .01), Suicide decedents without known MHP had significantly greater odds of other life stressors, such as a?criminal legal problems (ac-R 1.7, 95% CI or evictionfloss of home 1.4, 95% CI They had significantly lower odds of recent release from any institution, but when a release was indicated, they Were significantly more likely to be released from a correctional facility (ath 4.5, 95% CI or hospital 1.3, 95% CI Among decedents with known MHP who were recently released from an institution, from institution . Those with known MHP also more frequently had job 42.8% andfor financial problems (16.8% vs. .05). Decedents without known MHP had significantly greater odds of a recentfimpending crisis 1.4, 95% CI When the type of crisis was known, it was most frequently related to an intimate partner (35.2%l, physical health criminal legal family relationships or a job problem. Over one-fourth of decedents with a known also had recent or impending crises, most frequently related to an intimate partner physical health or relationship pecedents without known MHP had signi?cantly greater odds of criminal legal 1.6, 95% CI and significantly lowered odds of job-related (alei 0.1, 95% CI 0.5- family 0.3) Suicide decedent's-without known significantly great-or odds of leaving a suicide note 1.2, 95% CI while decedents with known MHP more often had a history of suicidal ideation (40.8% vs. 23.9%, 5 .01] and attempts {29.4% vs. 10.3%, pg Conclusions and Comments (135]100 words) 1999-2016, age-adjusted suicide rates among people a 10 years increased 25.4% overall. Forty-four states saw significant rate increases, and one (Nevada) state saw a significant decline. Suicide ai_nong males. This signal of increasing Vulnerability to sgcide among of females towards-suicide aligns with and an annual increase of 18.8% per year in emergency department visits for self-inflicted injuries among youn females, aged of 10 and-14, ?1 the pEflOElfpri'Ll 2008-2015??3 These increases may hint at a narrowing of the 9 More research into this troubling suicide gender gap, historically weighted towards males trend is needed. had a known mental health preblereMHP. This group was challenged by comorbid substance abuse problems had a history of mental healshw and/or substance abuse treatment and over half were currently in treatme at the time of their deaths, much more support for this vulnerable population is needed. This includes the ne for broader implementation of affordable and evidence-based treatments, such as doctor-patient collaborati 3 rates increased by more than 25% in 30 states and upwards of 50%[n sorn?. Ameog?iemale??ates increased in 43 states a o_ g_f_e_n1a_l_e_s and W?ii? recent reports that-identifiedmdicating a 53% increase in middle?aged female suicide rates between 1999?2014 3 histories ofsuicide ideation and attempts While two-thirds of people with MHP nt Comment Can you say this? Wouldn't saying this imply that if IPP is checked as a precipitating factor, then relationship problems also has to be checked? Is that how coding of (PP works? I thought these two categories were mutually exclusive in that could be checked without relationship problems being checked. ?h This could be easily changed by just Lremoving "specifically? if we need to for space Comment [vili5]: Maybe we should be i more precise I[Cuttlment vid5 = Could probably cut this 1 I I I l' I ratio to save spaceOne important factor associated with suicide is Nearly half of people in this study ed ve a {Comment ICAI: Maybe could leave out Fiev 3.6.13 Draft 1 care models and cognitive-behavioral therapy?) Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. 1? While MHP are a significant contributor to suicide, 54% of suicide decedents in the-currentt?g study did not have a known . This group suffered more life stressors, especially related to relationships leg. intimate partner problems, arguments or conflicts, -recent perpetration of intimate partner violence}, but also related to other life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises [often related to the abovementioned factors); People with?known MHP also experienced life?arid other stressors apart from their . Th is group was more?likEIytoe?iiper?ience iob andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone. These may include: strengthening economic supports leg. housing stabilization policies, household financial support}, teaching coping and problem-solving skills and other pro? social norms, especially early in life to manage everyday stressors and teprevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments leg, reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote he p~seeking and positive social norms], and supporting people after a suicide has taken place to prevent survivors? risk and to assure safe reporting by the media in order to prevent suicide Comment vid51= Wondering if instead of recapping results, we could comment on the profound and not often discussed impact that life stressors, especially those that rise to the level of a crisis, can have. The research that shows the time amount of deliberation [which is often very low) may also be helpful to cite here, as these types of stressorsy?crises can represent short?term problems (vs. longer term Lproblems such as chronic mental illness) l' Comment lvidSI: Similarly, wondering if we could condense this instead of reiterating some of these results and instead comment on the concept of "functional impairment,? as a reason why people with mental health problems might have more jobffinancial problems leg, they often have more inconsistent I . . I . f??y Comment Ivid5]: Iknow we don?t want to really be adding words but I thought this is one thing that may not be obvious to everyone {why deaths of undetermined intent would affect suicide See what you think, and if we can't get the text in, maybe we could contagion.m at least three limitations. In four states, Maryland The findings Utah Massachusetts and Rhode Island state rankings might have been impacted by large proportions of deaths of undetermined intent+MQl,[ br by decreased percentages of undetermined deaths over time (UT, MA, Second, 1 Third, abstractors of data are limited to data . For example, medical and moneaiheath-hMH information are coronerx?medical included in investigative reports . Therefore, knowledge of the informant not captured directly from medical records but from examiner reports impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies{?] Suicide is a one of many associated factors. Resources such as Preventing Suicide: Technical Package of Policies, Programs, and Practices? and the can help states and communities prioritize comprehensive suicide prevention,sneak a footnote in Comment Is it nationally representative at this point? 0r no? J's. .Il Comment [vid5 : Some suggested rephrasing here, because i don?t want to trigger questions about why we would only use one data year of data. We have good reasons more states, and for the state level analyses we were looking at trends and therefore needed more years. However, wondering Comment IvidSI: Just to give our data enough credit, though, I am wondering if the 90% estimates could also be driven up by small or very selective samples. Was going to look into this but not sure what the reference is. Little wary of making it ksound like theirs is the gold standard. Rev 3.6.13 Draft 1 References BE UPDATED) 1 CDC. Web-based Injury Statistics Query and Reporting System (WISOARS). Atlanta, GA: US Department of Health and Human Services, 2016. 2 World Health Organization. Preventing Suicide: A Global Imperative. Geneva, Switzerland: World Health Organization; 2014. 3 Curtin, S.C., Warner, M., and Hedegaard, H.. "Increase in suicide in the United States, 1999-2014." (2015}. 4 Kegler SR, Stone DM, Holland KM. Trends in Suicide by Level of Urbanization United States, 1999-2015. MMWR Morb Mortal Wkly Rep 5 National Action Alliance for Suicide Prevention, Office of the Surgeon General. 2912 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: US Department of Health and Human Services, Office of the Surgeon General; 2012. 6 Davidson, L., Potter, L, and Ross, V. Surgeon General's Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 1999. 7 American Association. Diagnostic and Statistical Manual of Mental Disorders American Association; Arlington, VA. 2013. 8 Mercado MC, Holland K, Leemis RW, Stone DM, WangJ. Trends in emergency department visits for nonfatal self~inflicted injuries among youth aged 10 to 24 years in the united states, 29014015. JAMA 9 Canetto, 5.5., and Sakinofsky. l. "The gender paradox in suicide." Suicide and lifevthreotening behavior 1998: 1?23. 10 Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: a technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 2017. 11 Isometsa E. autopsy studies?a review. European 2001;16i7k379?385. Acknowledgments Conflict of Interest None Corresponding author Stone Tables and Figures Options for ref 11 - .Hawton K, van Heeringen K. Suicide. Lancet. 2009373: 132281. Doi.org/10.1016/50140- 67'36 06 6032-X Canetto SS, Sakinofsky I. The Gender Paradox in Suicide. Suicide Life-Threatening Behavior. 1998; 28:1-23. From Katie: 1 went through the latest draft vii/Alex?s comments, comparing the results and discussion. Most lined up, although there were a few places where I noticed some mismatch is from the draft, bold are my comments]: "From 19994016, age?adjusted suicide rates among people a 10 years increased 25.4% overall.? We give the national (overall) number only in the abstract, not in the results. Rey 3.6.13 Draft 1 "Suicide rates increased by more than 25% in 30 states and upwards of 50% in some?. We only mention one [North Dakota) that is over 50% in the results (rather than "some?}. ?While two-thirds of people with MHP had a history of mental health and/?or substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed.? The highlighted point is only in the table right now- could add to results text. "While are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MHP. This group suffered more life stressors, especially related to relationships (eg. intimate partner problems, arguments or conflicts, recent perpetration of intimate partner Didn?t notice this the first time around, but the highlighted portion should read "recent perpetration of interpersonal violence? that variable is not speci?c to IPU. ?People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead}. Rev 3.6.13 Draft 1 Structured abstract {241,{250 words) This is the key strati?cation in my view. Scott?s comment: Background: Suicide rates have been risking in the United States Examining state?level trends in, and I. contributing circumstances to, suicide; can inform comprehensive state suicide prevention planning. H'[Comlment SCOWS commerjti The li/[Cornment tg59 : I like Scott? 5 Methods. Trends In age-adjusted suicide rates, overalLane-by sex and statd,_ among people a_g_ed 2:10 _years, J, were eaic?ulaeeal-[ lbsin?data from the National Vital Statistics System. Changen's In rates and I?EComment Iiike this the way You state rankings were across six consecutive three year periods {1999- 2001, 2002- 2004, 2005- I i{Comment [tgs?Jlt ldon't think we need 2007, zoos-2010, son-2013, 2014-2o1s}. Data from [thepols from the National violent Death Reporting System, across 27 states, were analyzed to precipitating circumstances I: Comment lsnk?l: I suicide decedents with and without known mental health problems. I Results: Statistically significant were identified 44 states. for the U. S. ?[C0nfmem Should we 53V. overall and for 30 states individually,[ the study period] Peerage 3i." Scott 5 comment. - . I I "KL/{Comment Do you want to say WithOUt 1 :x I{Comment if we have another re_s?pe_c itvelyr respectively {Comment ?359]: We can?t focus on Comment Need to rephrase this ea-nt?rtbeting? . Comment vid?l= Could we shorten to 'rr 1 I Conclusions: Suicide rates have risen significantly' In the U. 5. and across most states from 1999-2016100 single i?I i ff; Comment like primary factor alone contributes to suicidel_ .i IrCumsiancedI out: i Iinge ti In. riff {Comment i think. a diet of people Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population?based I ?If; {comm?nt IHKU 3 ?mm" ?we 5 l' public health approach eteategiesbased_ on_ the be_s_t 1i?? Comment "359]: Ithink this frame multiple levels (individual, familyfrelationship, community, societal], focused onbreventing risk before it startsL _?fI'lg' identifying and supporting people already at risk preventing- reattempts, and caring fors supporting uryiv- If; Comment to change ,{Comment' [tg59l: Deb in the fact sheet i I I: {Comment Need to update att?eea?sulcide deco giant 9' aurviving loved n: it- INTRODUCTION BACKGROUND AND PURPOSE {210(250 words} TOTAL Comment Shouid' we add In 2016, nearly 45,000 suicides {13.4f100,000i occurred in the United States If?! {@menut [3110- Need reference. here. I: [my 0 ,ions. and females across racialfethnic groups and across urbanization nonfatal self- harm Injuries increased lay?more than 40% between 2001 and 2015. 5 In 2015, suicides and self? I,[Comment Probably could cut harm' Injuries cost the nation more than E369 billionin direct medical and work loss costs?l Suicide' Is rarely 551'}? caused by one factor; rather the risks are often Comment ISDG: Do we ?it? to familylrelationship, community, and societal. '5 Despite this prevention primarily centers on [mental illnesq?le. g. ., ?[Comment IFKMI yd to leave it dopression bipolar disorder; Other factors associated with suicide include social economic access people at Fail} 1 {Comment ?3591' ?Bree coping and problem? solving skills, team?seemeamw air-[Comment [tgsti]: We need to keep this a prior suicide attempt among others.l3 While Although the Surgeon a General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states I ?1e Shourd 1 Rey 3.6.13 Draft 1 struggle to make this a reality.? ,this study analyzes trends In To assist states state suIcIde rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. Comment ltngI: I feel like we need a x? paint like this to set up why we are examining those with and without mental health problems separately. Comment There might not be enough room for an additional reference, but might want to consider including one here to highlight the rationale for METHODS [259/250 words} tabulated based on National Vital Statistics System coded death certificate records {international Classification of Diseases 10'? Revision underlying-cause-of death codes KEG-X84, ?(310. Age-specific population estImates were obtalned from 5. Census Bureau I National Center for Health Statistics bridged-race population data releases. National and state?level suicide rate estimates were calculated for six consecutive three year aggregate perIods 5 ye?rslrom 1999-2016. Rate estimates were age-adjusted to the U5. year 2000 standard population and expressed per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes 4 System were used to deaths by suicide among decedents with and without known current mental health problems iaierawi health prooleinsfyi I-P are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence [captured ie?sepaisaee?yasg?rg?yiaelesl. eempi?ies?aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports [including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estImate adjusted odds ratios with 95% con?dence intervals controlling for age group, sex, and race! ethnicity. RESULTS [696l500 wordsl The most recent overall suicide rates (representing 2014-2015] ranged from 6.9 (District of Columbus to 29.2 {Montana} per 100,000 persons per year, a four?fold difference [Table Across the entire study period. rates increased in all but one state {Nevada}, with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage Increases in rates ranged from +5.51% {Delaware} to (North Dakota), with percentage Increases of at least 25% observed in over half of all states as well as nationally 1- "l uh, forMstates as T-L ?7i 3' r_?H Modeled suicide rate trends well as for the .5. overall (Table 1] By sex, rate trends .I _s .Ie in 34 states for males and In 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was +1 5% By sex, the national AAPC was for males and +10% for females Lei health problems were I Suicide decedents with and without iN=11,039i known compared While both groups were predominately male and non?Hispanic thte decedents wIthout known MHP had 2.3 greater odds of being male (95% confidence interval 2.2 2 and sIgnIfIcantly greater odds of being racialfethnic minorities [odds ratio range: 1.0-2.1; 95% CI range They also had 5IgnIficantly greater odds of perpetrating homicide?suicide iadiusted odds ratio {aCIFtl 2.0, 95% CI Lexcluding persons under 10 years old. Ftev 3.6.13 Draft 1 2.23.8}, of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the U.5. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more if than those without MHP [19.8% vs. most frequently by over-themounteri'otherwise unclassi?ed drugs opioids antidepressants or bentodiazepines All suicide decedents with known MHP and approximately 35% without had Decedents without known MHP had signi?cantly greater odds of any type of relationship problem 1.3, ft 55% CI and [specifically of intimate partner problemslaQB 95%Elf arguments; on - rid conflicts {aDFi 1.4, 535% CI and Hose?recently perpetrateis'i rigs-F interpersonal violence 2.0, 95% CI 1.57.2.4}. legal problems 1.7, 95% CI or evictiony?loss of home {51th 1.4, 95% Cl 1.24.6} had significantly lower odds of recent release from any institution, but when a release was indicated, they were more likely to be released from a correctional facility 4.5, 95% CI or hospital 1.3, 95% CI Among decedents with known MHP who were recently released from an institution 42.3% from a institution; Suicide decedents without known MHP had significantly greater odds of leaving a suicide note 1.2, 95% 5 .01] and attempts [29.4% vs. 10.3%, 5 Conclusions and Comments (2351200 words) fortunate-tar: r-im 1999-2016, orty?four states saw significant rate increasesL and one {Nevada} state saw a signi?cant decline. Sui rates increased by more than 25% in 30 states and upwards of 50%[n WEE {North Qiiligli} ii '1 'i I l' had significantly greater odds of other life stressors, such as escriminal CI while decedents with known MHP more often had a history of suicidal ideation {40.8% vs. 23.0%, Comment [tgsti]: Suggestion to avoid saying ?known" again. Can just use "information? Comment ltgsii}: Suggest moving this sentence here to improve the flow. Comment IHKII: Can you say this? Wouldn?t saying this imply that if IPP is checked as a precipitating factor. then relationship problems also has to be checked? Is that how coding of IPP works? I thought these two categories were mutually exclusive in that IPP could be checked without relationship problems being checked. Jr. This could be easily changed by just ,removing "specificaily? Comment ltngI: I suggest moving point here for flow. I think we can shorten it in this way to save on words and to focus on the fact that the most common crises applied to both groups. This helps ,set up our discussion. Comment [vid5]: Could probably cut this i if we need to for space Comment vid5 : Maybe we should be more precise '1 1 I ,i J's. h. omment Need to take out ?some" here because North Dakota is the only state over the 50% mark. cide Rev 3.6.13 Draft 1 Comment Maybe could leave out Lratio t0 SEUE space. I 1 Comment Itgsg?l: This paragraph could be shortened a lot. We could have tested the change Il'l the seat gap but we did not present this so the point seems tangential. also. we also don't talk about sex 1 F. differences in circumstances or prevention I strategies so this text could get the reader I wondering about issues that we don"t discuss One Important factor associated with suicide is mental health Nearly half of Comment vid5 = Wondering if instead had a known amntaliseeithqesoieemmri P, This group was challenged by comorbid 01? recapping WE comment substance abuse problems and histories of suicide ideation and attempts While two- 5 on the profound and not often . . impact that life stressors, especially those thirds of people with MHP had a history of mental healthivj? and/or substance abuse treatment and over half . . . . I that rise to the level of a cri5is, can have. were currently in treatment at the time of thew deaths support The research that shows the time amount . This includes broader implementation of of deliberation (which is often very low) may also be helpful to cite here, as these Ir' types of can represent short?term problems lvs. longer term A I i alt JK. this vulnerable population affordable and evrdence based treatments, such as doctor?patient collaborative care models and cognitive- problems such as chronic mental illness} I behavioral therapy ?2 Additionally, greater access to behavioral health providers especially ?1 underserved areas as Is healthcare systems change that supports suicrde prevention and patient safety 12 I Comment ltgs?JI: I think this is worth I considering. We have the space to include a concise point about how situational factors min contribute to immediate risk .. significant contributor to suicide, 54% of suicide decedents in we tori ontth Is study did not have a known Hi This group suffered moreiife stressors, especially related to relationships lag. intimate i partner problems, arguments or conflicts, recent perpetration of intimate partner violence], but also related to other life stressors such as crimina ~legal matters, evictionfloss of home, and recent or impending crises (often I r, related to the abovementioned factorsl] People] with known MHP also experienced life and other stressors [Comment [th9]: I like this sentence apart from their This group was more likely to experience iob andfor financial problems. Also common Comment lyidSI: Similarly, wondering If were intimate partner problems physical health problems and recent or impending crises we could condense this instead of reiterating some of these results and instead comment on the concept of I Itgs9l: The reader won?t know what] I ou mean about norms here. I know we don?t want Comment lvid?l: to really be adding words but I thought this is one thing that may not be obvious to everyone {why deaths of II I Comment I'm not sure that we need this. ifwe think it helps then i suggest that we shorten it and move it to the end to improve the flow. I added an ?l These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone strategies may include: strengthening economic supports leg. housing I l" JL. stabilization poliues household financial support}, teaching coping and problem-solving skills [and promoting social connectedness to increase a sense of belongingness and access to informational tangible emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments leg, reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help?seeking and positive social norms}, and-supporting people after a suicide has taken place to prevent survivors risk and toassuringe safe reporting by the media in order to prevent suicide if: Comment [tg59 : This was in the 2015 55. I think it would be good to contagion.12 Theo;- findings .- his re.? Ivi -.at least three limitations. In four states UT MA RI) state-rankings might have been impacted by ll" include it here too_ It will be good to large proportions of deaths of undetermined intent iiv?iDi con?rm the accuracy with Katie. hr by decreased percentages o__f undetermined deaths over time If ,rrCornment Is it nationally Second, bit: run: i- run. '-,ill representative atthis pointComment yid5]: Some suggested I rephrasing here, because i don?t want to trigger questions about why we would only use one data year of data. 4? lion-eh'ei - 4 Ftev 3.6.13 Draft 1 Third, abstractors of data are limited to data included in the-investigative reports they receive. For example, medical and mental?healthm information are not captured directly from medical records but from (rug. iamily, vii-i coronerfmedical examiner friends. Therefore. knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and 13 Suicide is a growing public health problem and mental illness is an important risk one of Resources such as Preventing Suicide: Technical Package of Policies, Programs, and Practices12 and data from the National ?vioient-Death can help states and communities prioritize comprehensive suicide prevention. References BE 1 Wlsqars fatal injury reports 2 World report 3 films Data brief 201:1 4 Kegler et al 201? 5 WISQARS Nonfatal injury reports 6 National Strategy for Suicide Prevention 7 Davidson. L., Potter. L., and Ross, v. {19991 Surgeon General?s Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 3 ICU-10 9 DEM-5 Melissa?s paper 11 Need ref ?12 Technical package Acknowledgments Con?ict of Interest None Carresponding author Stone Tables and Figures Options for ref 11 :Hawton K, van Heeringen K. Suicide. Lancet. 20091373: Bill-81. sissigsisosgj -. Canetto SS, Sakinofsky The Gender Paradox in Suicide. Suicide 8L Life-Threatening Behavior. 1998; 23:1-23. From Katie: the latest draft xiv/Alex?s comments; comparing the results and discussion. Most lined up, although there were a few places where I noticed some mlsmatchl is from the draft, bold are my comments}: "From 1999-2016, age-adjusted suicide rates among people a 10 years increased 25.4% overall." We give the national {overall} number only in the abstract, not in the results. 5 l' .I 1 I Comment ltgs?'l: Iagree with the .F Comment IvidSI: Just to lgive our data comment below that including the 90% estimate here could be problematic. It might be better to describe the alternative types of studies. I suggested an edit so you can see what i mean but this will need to be updated to be Lconsistent with what you cite. JL. enough credit, though, I am wondering if the Boss estimates could also be driven up by small or very selective samples. Was going to look into this but not sure what the reference is. Little wary of making it Lsound like theirs is the gold standard. Comment Request to change reference number 3 from the 2014 NCHS data brief to the more recent 201? MMWR that looked specifically at sex, racefethnicity, age group. mechaniSm by Lurbanlzation level. Rev 3.6.13 Draft 1 "Suicide rates increased by more than 25% in 30 states and upwards of 50% in some?. We only mention one (North Dakota) that is over 50% in the results (rather than "some"l. ?While two-thirds of people with MHP had a history of mental health and/?or substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed.? The highlighted point is only,I in the table right now- could add to results text. "While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MHP. This group suffered more life stressors, especially related to relationships intimate partner problems, arguments or conflicts, recent perpetration of intimate partner Didn?t notice this the first time around, but the highlighted portion should read "recent perpetration of interpersonal violence" that variable is not specific to IPV. ?People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job and/or financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead}. Rev 3.13.18 Draft 2.1 Short title: Vital Signs: increasing Trends in State Suicide Rates and Contributing Circumstances 2 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 3 Kristin M. Holland, PhD,?l Ashe Z. Ivey-Stephenson, Alex E. Crosby, MD1 4 Structured abstract [2119;" 25DI wordsl 5 Background: Overall suicide rates have been rising in the United States Examining staterlevel trends in suicide and the multiple contributing circumstances can inform comprehensive state suicide prevention 7 planning. it Methods: Trends in age?adjusted suicide rates. by state and sex, among people aged 210 years. were assessed 9 using data from the National Vital Statistics System. Changes in rates were examined across six consecutive l0 three-year periods from 1999-2015. The National 1violent Death Reporting System [2015}, covering 27 states, was used to ampere?Examine the precipitating circumstances amongbeeween suicide decedents with and [2 without known mental health problems l3 Results: Forty-four states saw statistically significant suicide rate increases over the period. In states, l4 rates increased by more. Male suicide rates increased 34 states while female rates increased signi?cantly in People with and without known MHP had both differing and [5 lo similar precipitating circumstances associated with their suicides. Many-Several circumstancefaetoss, such as [7 relationship problems {39.6 and life stressors and and recent crises (25.0 and IS respectively, were more likely among those without known MHP, but were common among both groups. 19 Conclusions: Suicide rates rose significantly across most states from 1999-2016. Varied circumstances beyond 20 l?leF alone contributed to suicides among people with and without known lleP. 22 based on the best available evidence to prevent s'iislnpiesuicide risks before they occur, identify and support 23 people already at risk, prevent?reattempts, and help friendsifamily after a suicide occurs. 24 INTRODUCTION 25 BACKGROUND AND PURPOSE {2431'250 words] 2? 2016, suicide rates increased among males and females, across racial/ethnic groups. and across urbanization 2 Implications for Public Health Practice: States can use a comprehensive 28 levels??31 Suicide?js the 10?1 leading cause of death and is among the only leading causes to be increasing?" 29 Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, 30 increased more than 40% between 2001 and 2015.? Together, suicides and self-harm injuries cost the nation 3 more than $69 billion in direct medical and work loss costs.1 33 reaching across the social ecology individual, familyfrelationship, community, and societal levels]. Such an 34 approach highlights that suicide is rarely caused by any single factor but rather is Despite 35 this, suicide prevention efforts largely focus on identifying and treating high?risk individuals with mental illness.?i 36 However, other associated risk factors exist and include social and economic problems. access to lethal means 3? leg. substances, firearms} among people at risk, poor coping and problem-solving skills, and prior suicide 38 attempts, among others? Expanded awareness of the multiple circumstances that contribute to suicide risk ?1 Comment ltgs'?l: Suggestion because we have not explained what we mean by multi?level yet. Also, we use ?comprehensive? in the background so - Lwe can be consistent. Comment snkti = It?s 15.5 [age?adjusted} when using the population denominator 2 1o years of age. That?s consistent with how we present rates elsewhere in this report (and with what we've already stated in 32 The National Strategy for Suicide Prevention5 calls for a public health approach to suicide with prevention efforts _the Abstract]. {Comment Isnk?l: Extra mark? I 39 apart from mental health problems alone, can help 1 a. Rev 3.13.13 Draft 2.1 40 states in reaching this goal through and?toward?comprehensive suicide prevention activitiesplaaning, this study _t:"mment lVid5li Did we want to say 41 analyzes state?specific trends in suicide rates, assesses the multiple factors associated with suicide, and provides y? why? This seems to alwals come uE. .?ll Comment Isnk?l: 42 prevention recommendations. if Need to be careful here no longer true 43 METHODS [246,250 words} nationally with the newly adopted 30% 44 Suicide rate estimates and trend analyses exclude data for personsle? years :l threshold. 45 tabulated based on National Vital StatistiCs System coded death certificate records {lnremotionoi Clossb?r'cotion of 4a Diseases 10?? Revision underlying-cause-of death codes XEO-X84, Y87.0, U03). Age?specific population 4? estimates were obtained from US. Census Bureau/National Center for Health Statistics bridged?race population 48 data releases. 49 50 National and state-level suicide rate estimates were calculated for sh: consecutive three-year aggregate periods 5] from 1999-2016. Rate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed per 52 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data 53 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 54 are reported in terms of average annual percentage changes 55 56 Data from 2015 from the 27' states with complete data participating in the National Violent Death Reporting i i i I 5? System were used to compare the characteristics among suicide decedents with and without known 58 current mental health problems are defined in as disorders and listed in the 59 Diagnostic and Statistical Manual of Mental Disorders. Fifth Edition (ow?51,? with the exception of alcohol and i 60 other substance dependence (captured separately). aggregates data from three primary data sources: 5 6 death certificates, coronerfmedical Examiner reports {including toxicology), and law enforcement 62 reports. Decedents with and without known MHP were compared using Chi?square tests; logistic regression was i 63 used to estimate adjusted odds ratios with 55% confidence intervals controlling for age group, sex, and 64 racei?ethnicity. i as nesucrs (searsno words] i as The most recent DVErall suicide rates [representing 2014?2015] ranged from 6.9 (District of Columbia] to 29.2 i at {Montana} per 100,000 persons per year, a four-fold difference (Table Across the entire study period. rates 5' 68 increased in all but one state {Nevada}, with increases ranging from +0.2 (Delaware) to +8.1 {Wyoming}. i :59 Percentage increases in rates ranged from +59% {Delaware} to (North Dakota], with percentage ?0 increases of at least gee-303:1- observed in arise-sehalf of all Tl T2 Modeled suicide rate trends indicated significant increases for 44 states. as well as for the U5. overall {Table 7?3 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. 7'4 Nationally, the model-estimated AAPC for LIE-overall suicide rates was By sex, the national RAPE was 75 +11% for males and +16% for females. 76 77' Suicide decedents with {N=5,402l and without known MHP were compared. While both groups were Til predominately male and non?Hispanic white, suicide decedents without known MHP had 2.3 greater odds of 79 being male [95% confidence interval and significantly greater odds of being raciali'ethnic 80 minorities (odds ratio range: Cl range They also had significantly greater til odds of perpetrating homicide-suicide (adjusted odds ratio [aDRl 2.9, 95% CI of firearm suicide 82 LE, 95% CI and of pesitive toxicology results for alcohol iaDR 1.2, 95% Cl Fifteen 33 percent of those with known MHP and 20% without ever served in the U5. military. Ftev 3.13.18 Draft 2.1 84 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more 85 than those without MHF {19.8% vs. most frequently by over-the?counterfotherwise unclassi?ed drugs as {35.8%l, opioids antidepressants or henzodiazepines 8? All suicide decedents with known MHP and approximately 35% without had precipitating 88 circumstances information. People with MHF were more likely to have any substance abuse problems (31.5% vs I 89 25% 01}. While two-thirds ofthose with known MHP had a history ofMH or substance abuse treatment 90 just over half [54.11%] were in current mental health treatment at the time of their deaths4i, specifically intimate partner problems laDRz arguments/conflicts 93 1 4, 95% CI and recently perpetrating interpersonal violence laUR 2.They also had significantly greater odds of other life stressors, such as criminal legal problems laDR 1.11, 95% CI I 95 1 5-19} or evictionfloss of home (309. 1.4, 95% CI and they were also more likely to have Dada 96 crisis within the preceding or upcoming two weeks 1.4, 95% CI Among both groups, the most 97 common crises were intimate partner and physical health problems. 93 Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but when a 99 release was indicated, they were more likely to he recently released from a correctional facility 4.5, 95% IUD CI 3.2-6.3} or hospital laUR 1.3, 95% CI Among decedents with known MHP warmers?recently IOI released from an institution 42.3% were released from facilities. lil2 Suicide decedents without known MHP had significantly greater odds of leaving. a suicide note laDR: 1. 2, 93% 03 l04 IUS Conclusions and Comments [6931'100 words] Cl: [Go From 1999? 2015, 44 states saw significant rate increases. Half of the states experienced increases of 30% or llJI'ir more." - - - -. - Rates increased significantly in 34 states among males and W3 states among females. [09 More research into the causes of these trends isfecessa? 110 One important factor associated with suicide is MHP. Nearly half of suicide decedents in had a known MHP. This group was challenged by comorbid substance abuse problems and histories of suicidale l2 ideation and attempts While two?thirds of people with MHP had a history of MH andfor l3 substance abuse treatment and over half were currently in treatment at the time of their deaths, additional I I4 support could help address the needs of this vulnerable population. This includes broader implementation of [5 affordable and evidence?based treatments, such as doctor-patient collaborative care models and cognitive- llo behavioral therapy.?3 Additionally, greater access to behavioral health providers, especially in underserved areas I is important, as is healthcare systems change that supports suicide prevention and patient safety through care llii transitions. 3 ll!) Wh'le l?leP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known l2? MHP. This groUp suffered more relationship problems and life stressors such as criminalrlegal matters, [21 evictionfloss of home, and recent or impending crises. This is noteworthy in light of findings that suggest many l22 suicides and attempts occur with minimal deliberation time, particularly among people without mental health 123 disorders and who faced impending life crises?"l? People with known MHP also experienced othermultiple life [24 and other stressors such as job andz?or ?nancial problems, relationship problems, and physical health problems 3 Comment ligs'?l: I'm wondering lfwe could add the 56's. This would make this consistent with the abstract and be more in line with our emphasis on the fact that there were differences but the circumstances are relevant for both groups. The structure would stay the same. We could use the approach that you used below [green highlighted text) to make the 9% comparison and reflect the signi?cance testing without giving the am and 95% Cl?s. This would also save :words. JL Comment lsnk?l: This sounds so precise all ofa sudden, compared to the very general benchmark of 25% in this same sentence. I'm thinking about a different Way to communicate this, but I don?t have anything better at the moment. . Iagree. Whatabout . takingoutthe ND point 1 Comment Itgsl?t lthink It would be good to cite some work that describes potential explanations and the need for additional research. Rev 3.13.18 Draft 2.1 [25 These findings point to the need to both prevent the conditions associated with mental health problems in the [26 first place and the need to support people with MHP to decrease their vulnerabilitv to poor social, health, and [2?3l economic outcomes.11 [28 These underscore the importance of state suicide prevention [29 activities that goes bevond a focus an NIH treatment alone. Prevention strategies may include: strengthening [30 economic supports le.g. housing stabilization policies, household financial support], teaching coping and problem-solving skills, especiallv early in life to manage evervdav stressors and prevent future relationship [32 problems; and promoting social connectedness to increase a sense of belongingness and access to I33 informational, tangible, emotional, and social support, as needed. Other strategies indicated lav these results [34 include creating protective environments le.g., reducing access to lethal means among people at risk, creating [35 organizational and workplace policies to promote help- seeking and positive social norms}, supporting people [36 after a suicide has taken place to?zrevent survivors has," and assuring safe r_e_po_r_t_in_g_ _bvm the In order to prevent suicide contagion la?ilil'hile few states have had the opportunitv for such a comprehensive approach, [38 states such as Colorado are taking up the challenge ferr?I??-IH __include some unrecognized suicided. [42 Second, Is not vet nationally representative. This studv used the most current data availablewhichsand [43 includes 27 states that represent half of the U.S. population. Third, abstractors of data are [44 limited to data included in investigative reports. For example, medical and MH information are not captured [45 directly from medical records but from key informants leg, familv, friendsl via coronerfmedical examiner I46 reports. Therefore, knowledge of the informant impacts completeness and accuracyr of the information reported, and In- depth studies with familv- members often see greater attributions to mental health and [48 substancL abuse; dis?orderslikely that some [50 people without known mental health problems' In the current studv were experiencing mental health challenges at the time ofdeath, diet were either not known or reported lav informants, or were not captured in reports [52 from primarv data sources. [53 [55 one of many. Resources such as Preventing Suicide: a Technical Package ofPolI'oes, Programs, and [So Practices? and data from the can help states and communities prioritize comprehensive suicide prevention. [58 Acknowledgments [59 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital [60 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. [o2 Con?ict of Interest No conflicts of interest were reported. [63 [64 Corresponding author: Deborah M. Stone, dston93@cdc.gov [65 [66 Author Affiliations: Comment IvidSI: Still wondering if :this? ,_prevention." Suggest ordering alphabeticallv. Ljust ttveak it a little and would be fine] Comment [Vid5]: Responding to Tom?s people might get confused re: the term "sunrivor" is there a different waiiI to put Comment [tgsli]: You include two #12 references below. I not sure that buv the lack of opportunity frame here. What about saving "Some states, such as CD, have already developed and are implement plans for comprehensive suicide Comment Isnk?l: J5. .ullh. Comment vid5]: Isn?t this referring to the ?rst part {large proportions of undetermined deaths]? I like that we're explaining this point a little but the structure here kind of looks like We?re referring to the decreased 9-65 [izirobabliiI Comment [tgs?h This makes it sound like the and these studies use similar approaches. It looks to me like the 90% estimate in the review you cite includes mental or substance abuse disorders. What do you think of the edit suggested? i?m still not sure of the best way to make this point but I think this is Lclose. ?4 email from earlier: MHP in are based on reports that the decedent had a MH disorder that is a diagnosis in the DSM. Often informants will report 5.: I I-Im-u- l-n. -nI-I?Innl- Comment [tgs9 : Did our classi?cation relv on a diagnosis? This implies that we ,_did. Comment lvid5]: MHP in are based on reports that the decedent had a MH disorder that is a diagnosis in the DSM. Often informants will report information that seems to reflect that the 1' I Comment ?359]: I think some will take Jssue with saving it is "just? one of many. 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 I85 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 Rev 3.13.18 Draft 2.1 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References 1 2 10 11 12 12 CDC. Web-based Injury Statistics Query and Reporting System (WISOARS). Atlanta, GA: US Department of Health and Human Services, 2016. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca M1. Suicide Trends Among and Within Urbanization Levels by Sex, Raceg?Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ Kegler SR, Stone DM, Holland KM. Trends in Suicide by Level of Urbanization - United States, 1999-2015. MMWR Morb Mortal Wkly Rep Kochanek KD, Murphy SL, Xu J, Arias E. Mortality in the United States. NCHS Data Brief 2017;293. US Department of Health and Human Services. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: 2012. Rosenman, Stephen J. "Preventing suicide: what will work and what will not." The Medicaliournoi of Australia 169.2 [1998): 100-102. American Association. Diagnostic and Statistical Manual of Mental Disorders American Association; Arlington, VA. 2013. Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: a technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 2017. Simon, Thomas R., et al. "Characteristics of impulsive suicide attempts and attempts-rs." Suicide and Life Threatening Behavior 32.51 {2002}: 49-59. Schiff, Lara B., et al. ?Acute and chronic risk preceding suicidal crises among middlevaged men without known mental health andfor Substance abuse problems: An exploratory mixed~methods analysis." Crisis: The Journal of Crisis intervention and Suicide Prevention 36.5 (2015): 304. World Health Organization. Risks to Mental Health: An Overview of Vulnerability and Risk Factors. August 2012. Caine, Eric 0., et al. "Comprehensive, integrated approaches to suicide prevention: practical guidance.? injury prevention injuryprev?201?i. (epub ahead of print) Cavanagh, J.T.O., Carson, A.1., Sharpe, M., Lawrie, S.M. autopsy studies of suicide: A systematic review. Medicine, 2003; 33, 395-405. Tables and Figures {attachments} Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 178071800 Flev 3.13.18 Draft 2.1 I Short title: Vital Signs: increasing Trends in State Suicide Rates and Contributing Circumstances 2 Deborah M. Stone, Sclil;I Thomas R. Simon Phi];1 Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 3 Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 4 Structured abstract [2491250 wordsl 5 Background: Dverall suicide rates have been rising in the United States lU.S.l. Examining state?level trends in a suicide and the multiple contributing circumstances can inform comprehensive state suicide prevention 7 planning. 8 Methods: Trends in age?adjusted suicide rates. by state and sex, among people aged all] years, were assessed 9 using data from the National Vital Statistics System. Changes in rates were examined across six consecutive IO three-year periods from 1999-2015. The National 1violent Death Reporting System l2D15l, covering 27 states. II was used to compare the precipitating circumstances between suicide decedents with and without known [2 mental health problems l3 Results: Forty-four states saw statistically signi?cant suicide rate increases over the Moeriod. In 30 states, I4 rates increased by 25% or more. Male suicide rates increased significantly in 34 states while female rates [5 increased slg??yin 43. People with and without known MHP had both differing and similar lo precipitating circumstances associated with their suicides. Many factors, such as relationship problems {39.6 and life stressors and and recent crises {26.0 and respectively, were more likely IS among those without known l'leP, but were common among both groups. 19 Conclusions: Suicide rates rose significantly across most states from 1999-2016. Varied circumstances beyond 20 MHF alone contributed to suicides among people with and without known MHP. 2 Implications for Public Health Practice: States can use a multl-level public health approach based on the best 22 available evidence to prevent multiple suicide risks before they occur, identify and support people already at 23 risk. prevent-reattempts, and help after a suicide occurs. 24 INTRODUCTION 25 BACKGROUND AND PURPOSE {2431'250 words] 2? 2016, suicide rates increased among males and females, across racial/ethnic groups, and across urbanization 28 levels??3 Suicide is the 10?? leading cause of death and is among the only leading causes to be immeasing.it1 29 Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, 30 increased more than 40% between 2001 and 2015.? Together, suicides and self-harm injuries cost the nation 3 more than $69 billion in direct medical and work loss costs.1 32 The National Strategy for Suicide calls for a public health approach to suicide with prevention efforts 33 reaching across the Social ecology individual, familyx?relationship, community, and societal levels]. Such an 34 approach highlights that suicide is rarely caused by any single factor but rather is multi?deternnined.5 Despite 35 this, suicide prevention efforts largely focus on identifying and treating high~risk lndividuals with mental illness!? 36 However, other associated risk factors exist and include social and economic problems, access to lethal means leg. substances, firearms} among people at risk, poor coping and problem-solving skills, and prior suicide 38 attempts, among others? Expanded awareness of the multiple Circumstances that contribute to suicide risk 39 apart from mental health problems alone, can help achieve substantial reductions in suicide rates. To assist Comment lsnk?l: lt?s 15.5 [age-adjusted} when using the population denominator 2 1D years of age. That's consistent with how we present rates elsewhere in this report land with what we've already stated in the Abstract}. Rev 3.13.18 Draft 2.1 states in this goal and toward comprehensive suicide prevention planning, this study analyzes state-specific trends in suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METHODS (245;st words) Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (lnternationai Classification of Diseases 10? Revision underlying-cause-of death codes X60-X84, Y87.0, U03). Age-specific population estimates were obtained from U.5. Census Bu reau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods from 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and extpressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCS). Data from 2015 from the 2? states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics among suicide decedents with and without known current mental health problems (MHP). MHP are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured separately). aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racefethnicity. RESULTS (5931'600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia) to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +3.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to +5.16% (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rate trends indicated significant increases for 44 states, as well as for the US. overall (Table 1). By sex, modeled rate trends indicated significant increases in 34 states for males and i143 states for females. Nationally, the model-estimated AAPC for th_eoverall suicide ratesr was By sex, the national AAPC was +11% for males and +26% for females. Suicide decedents with and without known MHP were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval and significantly greater odds of being racial/ethnic minorities {odds ratio range: 102.1; 95% Cl range [1.04.3] They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the US. military. Ftev 3.13.18 Draft 2.1 84 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more Comment [snug]: This sounds so precise all ofa sudden, 85 than those without lleF {19.3% vs. most frequently by over?the?counterfotherwise unclassified drugs compared to the very general benchmark to opioids antidepressants or benzodiazepines of 25% in this same sentence. Pm thinking about a different way to i i ii? All suicide decedents with known MHP and approximately 35% without had precipitating 88 circumstances information. People with l'v'lHF were more likely to have any substance abuse problems (31.5% vs. communicate tl"ls- bu? don?t have i anything better at the moment. 39 25%, While two-thirds had a history of MH or substance abuse treatment {512964. just over half 9i} were In current mental health treatment at the time of their deaths. 1 9 Decedents without known MHP had signi?cantly greater odds of any type of relationship problem laDR 1.3, 92 55% CI specifically intimate partner problems 1.4, 95% CI iaDR 1.4, 95% CI and recently perpetrating interpersonal violence 2.0, 95% Cl They also i 93 94 had signi?cantly greater odds of other life stressors, such as criminal legal problems iaUR 1.2, 95% CI 1.5?1.9} 95 or evictioni?loss of home 1.4. 95% CI and they were also more likely to have a crisis within the 96 preceding or upcoming two weeks (30R 1.4, 55% Cl Among both groups, the most common crises 9? were intimate partner and physical health problemsDecedents without known MHP had signi?cantly lower odds of recent release from any institution, but when a 99 release was indicated, they were more likely to be released from a correctional facility 4.5, 55% CI 3.2- IUU 6.3) or hospital [ac-Fl 1.3, 95% Cl Among decedents with known MHP who were recently released IOI from an institution 42.8% were released from facilities. l?Z Suicide decedents without known had significantly greater odds of leaving a suicide note laDR 1.2, 55% [03 CI while decedents with known MHP more often had a history of suicidal ideation {40.8% vs. 23.0%, i 5 HM .01] and attempts [25.4% vs. 10.3%, 5.01}. i IUS Eonclusidns and Comments (EQBITDD words) i me From 1999?2015, 44 states saw significant rate increases. Suicide rates increased by more than 25% in 30 states ,l l?i? ?nd upwards of Swain 21E: state, Earth Dakota. Rates increased among males and IOR increased significantly in 43 states among females. More research into the causes of these trends is necessary. [09 One important factor associated with suicide is MHP. Nearly half of suicide decedents in had a known Ill) MHF. This group was challenged by comorbid substance abuse problems and histories of suicide Ill ideation and attempts While two-thirds of people with MHP had a history of MH andfor l2 substance abuse treatment and over half were currently in treatment at the time of their deaths, additional ?3 support could help address the needs of this vulnerable population. This includes broader implementation of I I4 affordable and evidencebased treatments, such as doctor-patient collaborative care models and cognitive [5 behavioral therapy.B Additionally, greater access to behavioral health providers, especially in underserved areas lie is important, as is healthcare systems change that supports suicide prevention and patient safety through care I transitions.El [Iii While MHP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known I 19 iv'lHP. This group suffered more relationship problems and life stressors such as criminal?legal matters, l2ii evictionfloss of home, and recent or impending crises. This is noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation time, particularly among people without mental health l22 disorders and who faced impending life People with known MHP also experienced multiple life and 123 other stressors such as job andforfinancial problems, relationship problems, and physical health problems. [24 Those findings point to the need to both prevent the conditions associated with mental health problems in the 3 Rev 3.13.18 Draft 2.1 [25 first place and the need to support people with MHP to decrease their vulnerability to poor social, health, and [26 economic outcomes.? [21r These results point to the need for comprehensive state suicide prevention that goes beyond a focus on Mt! [28 treatment alone. Prevention strategies may include: strengthening economic supports leg. housing stabilization [29 policies. household financial support}, teaching coping and problem?solving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as [32 needed. Dther strategies indicated by these results include creating protective environments reducing I33 access to lethal means among people at risk, creating organizational and workplace policies to promote help- 34 seeking and positive social norms}, supporting people after a suicide has taken place to prevent survivors? risk, [35 and assuring safe reporting by the media in order to prevent suicide contagion.a While few states have had the [36 opportunity for such a comprehensive approach, states such as Colorado are taking up the challenge.? [37? IheseThe study findings have at least three limitations. in four states MA RI rankings [38 might have been impacted by large proportions of Mdeaths of undetermined intent, or by decreased [39 percentages of endetesna?ined?suc?deaths overtime, which likely some unrecognized suicides. [40 Second, is not yet nationally representative. This study used the most current data available and includes 27 states that represent half of the U.S. population. Third, abstractors of data are [42 limited to data included in investigative reports. For example, medical and MH information are not captured [43 directly from medical records but from key informants leg, family, friends} via coronerlmedical examiner [44 reports. Therefore, knowledge of the informant impacts completeness and accuracy of the information [45 reported. This may explain some of the discrepancy between the prevalence of MHP reported here and studies I46 that obtain estimates based on in-depth interviews with It is likely that some people without [47r known mental health problems in the current study were experiencing mental health challenges at the time of [48 death, but the absence of a diagnosis underscores the importance of addressing the range of other contributing I49 circumstances. [50 Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many. Resources such as Preventing Suicide: a Technicai Package of Policies, Programs, and [52 Practices? and data from the can help states and communities prioritize comprehensive suicide [53 prevention. [54 Acknowledgments [55 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital [56 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. [57 Con?ict of Interest No conflicts of interest Were reported. [59 [60 Corresponding author: Deborah M. Stone, dstone3@cdc.gov amass-3942 [(52 Author Affiliations: [63 1Division of Violence Prevention, National Center for Injury Prevention and Control, z[Zlivision of Analysis, [64 Research, and Practice Integration, National Center for Injury Prevention and Control, CDC I as References Comment [snk?h Su est ordering alphabetically166 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 Rev 3.13.18 Draft 2CDC. Web-based Injury Statistics Query and Reporting System Atlanta, GA: US Department of Health and Human Services, 2016. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ Kegler SR, Stone DM, Holland KM. Trends in Suicide by Level of Urbanization - United States, 1999-2015. MMWR Morb Mortal Wkly Rep 2017;6611011270-273. Kochanek K0, Murphy SL, Xu J, Arias E. Mortality in the United States. NCHS Data Brief 2017:2513. US Department of Health and Human Services. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: 2012. Rosen man, Stephen J. "Preventing suicide: what will work and what will not.? The of Aastraiia 169.2 [1998}: 100-102. American Association. Diagnostic and Statistical Manual of Mental Disorders American Association; Arlington, VA. 2013. Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: 3 technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 2017. Simon, Thomas R., et al. "Characteristics of impulsive suicide attempts and attempters.? Suicide and Life? Threatening Behavior 32.51 {2002}: 49-59. Schiff, Lara B., et al. ?Acute and chronic risk preceding suicidal crises among middle-aged men without known mental health andfor substance abuse problems: An exploratory mixed-methods analysis.? Crisis: The iournoi of Crisis intervention and Suicide Prevention 36.5 (2015): 304. World Health Organization. Risks to Mental Health: An Overview of Vulnerability and Risk Factors. August 2012. Caine, Eric 0., et al. "Comprehensive, integrated approaches to suicide prevention: practical guidance.? injury prevention [2017): injuryprevpzoli?. (epub ahead of print) Cavanagh, J.T.O.. Carson, A.J., Sharpe, M., 8: Lawrie. S.M. autopsy studies of suicide: A systematic review. Medicine, 2003; 33, 395-405. Tables and Figures {attachments} Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 1780/1800 UJM Rev 3.13.18 Draft 2.1 Short title: Vital Signs: Increasing Trends in State Suicide Rates and Contributing Circumstances Deborah M. Stone, Sclil'l;1 Thomas R. Simon PhD,-l Katherine A. Fowler, PhD;l Scott R. Kegler, Keming Yuan, Kristin M. Holland, PhD,l Asha Z. lvey-Stephenson, Alex E. Crosby, I?lel Structured abstract [2491250 words) Background: Overall suicide rates have been rising in the United States Examining state-level trends in suicide and the multiple contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates, by state and sex, among people aged 210 years, were assessed using data from the National Vital Statistics System. Changes in rates were examined across six consecutive three-year periods from 1999?2016. The National Violent Death Reporting System [2015), covering 27 states, was used to compare the precipitating circumstances between suicide decedents with and without known mental health problems Results: Forty-four states saw statistically significant suicide rate increases over the period. In 30 states, rates increased by 25% or more. Male suicide rates increased in 34 states while female rates increased in 43. People with and without known MHP had both differing and similar precipitating circumstances associated with their suicides. Many factors, such as relationship problems [39.6 and life stressors and and recent crises (26.0 and respectively, were more likely among those without known MHP, but were common among both groups. Conclusions: Suicide rates rose significantly across most states from 19992015. Varied circumstances beyond MHP alone contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a multi?Ievel public health approach based on the best available evidence to prevent multiple suicide risks before they occur, identify and support people already at risk, prevent-reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {243/250 words) In 2016, nearly 45,000 suicides {13.4f100,000) occurred in the United States Between 1999 and 2016, suicide rates increased among males and females, across racialg?ethnic groups, and across urbanization levels};3 Suicide, is the 101th leading cause of death and is among the only leading causes to be incnaasing.1"1 Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, increased more than 40% between 2001 and 2015.1 Together, suicides and self+harm injuries cost the nation more than $69 billion in direct medical and work loss costs.I The National Strategy for Suicide Prevention5 calls for a public health approach to suicide with prevention efforts reaching across the social ecology individual, familyfrelationship, community, and societal levels). Such an approach highlights that suicide is rarely caused by any single factor but rather is multi-determined.5 Despite this, suicide prevention efforts largely focus on identifying and treating high-risk individuals with mental illness.E However, other associated risk factors exist and include social and economic problems, access to lethal means substances, firearms) among people at risk, poor coping and problem~solving skills, and prior suicide attempts, among others.5 Expanded awareness of the multiple circumstances that contribute to suicide risk apart from mental health problems alone, can help achieve substantial reductions in suicide rates. To assist Rev 3.13.18 Draft 2.1 states in this goal and toward comprehensive suicide prevention planning, this study analyzes state-specific trends in suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METHODS (246f250 words] Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Classification of Diseases 10? Revision underlying-cause-of death codes X60-X84, Y87.0, U03). Age-specific population estimates were obtained from U.S. Census Bu reau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods from 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics among suicide decedents with and without known current mental health problems (MHP). MHP are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured separately). aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and raceiethnicity. RESULTS (593/600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia] to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 {Delawa re) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to (North Dakota), with percentage increases of at least 25% observed in over half of all states as well as nationally. Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall (Table By sex, modeled rate trends indicated significant increases in 341 states for males and d3 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known MHP were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval 222.5), and significantly greater odds of being racial/ethnic minorities (odds ratio range: 1.0?2.1; 95% CI range [16720]). They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the U.S. military. 100 101 102 103 104 105 106 107r 108 109 110 111 112 113 114 115 116 11? 118 119 120 121 122 123 124 Rev 3.13.18 Draft 2.1 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.8% vs. most frequently by unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without had precipitating circumstances information. People with MHP were more likely to have any substance abuse problems (31.6% vs. 25%, (.011. While two-thirds had a history of NIH or substance abuse treatment just over half were in current mental health treatment at the time of their deaths. Decedents without known MHP had significantly greater odds of any type of relationship problem 1.3, 95% C1 specifically intimate partner problems 1.4, 95% CI 1.4, 95% CI and recently perpetrating interpersonal violence 2.0, 95% CI They also had significantly greater odds of other life stressors, such as criminal legal problems 1.2, 95% CI 1.5-1.9) or eviction/loss of home 1.4, 95% CI 1.21.151, and they were also more likely to have a crisis within the preceding or upcoming two weeks 1.4, 95% CI Among both groups, the most common crises were intimate partner and physical health problems. Decedents without known MHP had significantly lower odds of recent release from any institution, but when a release was indicated, they were more likely to be released from a correctional facility 4.5, 95% CI 3.2- 6.3) or hospital 1.3, 95% CI Among decedents with known MHP who were recently released from an institution 42.3% were released from facilities. Suicide decedents without known MHP had significantly greater odds of leaving a suicide note 1.2, 95% CI 1.14.2}, while decedents with known MHP more often had a history of suicidal ideation (40.8% vs. 23.0%, ,o 5 .01} and attempts (29.4% vs. 10.3%, 5 .01). Conclusions and Comments (698.1700 words) From 1999-2016, 44 states saw significant rate increases. Suicide rates increased by more than 25% in 30 states and upwards of 52% in one state, North Dakota. Rates increased in 34 states among males and increased in 43 states among females. More research into the causes of these trends is necessary. One important factor associated with suicide is MHP. Nearly half ofsuicide decedents in had a known MHP. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two-thirds of people with MHP had a history of MH andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, additional support could help address the needs of this vulnerable population. This includes broader implementation of affordable and evidencevbased treatments, such as doctor?patient collaborative care models and cognitive behavioral thera py.E Additionally, greater access to behavioral health providers, especially in underserved areas is important, as is healthcare systems change that supports suicide prevention and patient safety through care transitions. 3 While MHP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a knOWn MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. This is noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation time, particularly among people without mental health disorders and who faced impending life People with known MHP also experienced multiple life and other stressors such as job andfor financial problems, relationship problems, and physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the 3 125 126 12?ir 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 145 146 147 148 149 .156 151 152 153 154 155 156 157 158 159 161 162 163 164 .165 166 Rev 3.13.18 Draft 2.1 first place and the need to support people with MHP to decrease their vulnerability to poor social, health, and economic outcomes.11 These results point to the need for comprehensive state suicide prevention that goes beyond a focus on MH treatment alone. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}, teaching coping and problem~solving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help- seeking and positive social norms}, supporting people after a suicide has taken place to prevent survivors? risk, and assuring safe reporting by the media in order to prevent suicide contagion.8 While few states have had the Opportunity for such a comprehensive approach, states such as Colorado are taking up the challenge.12 These findings have at least three limitations. In four states (MD, UT, MA, Rl), rankings might have been impacted by large proportions of deaths of undetermined intent, or by decreased percentages of undetermined deaths over time, which likely reflect some unrecognized suicides. Second, is not yet nationally representative. This study used the most current data available and includes 27 states that represent half of the U.5. population. Third, abstractors of data are limited to data included in investigative reports. For example, medical and MH information are not captured directly from medical records but from key informants family, friends) via coroner/medical examiner reports. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of MHP reported here and studies that obtain estimates based on inadepth interviews with It is likely that some people without known mental health problems in the current study were experiencing mental health challenges at the time of death, but the absence of a diagnosis underscores the importance of addressing the range of other contributing circumstances. Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many. Resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices12 and data from the can help states and communities prioritize comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, US Department of Health and Human Services, 2016. 4 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 l87 188 189 190 191 192 193 194 195 196 197 198 Rev 3.13.18 Draft 2.1 2 10 11 12 12 Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ Kegler SR, Stone DM, Holland KM. Trends in Suicide by Level of Urbanization - United States, 1999-2015. MMWR Morb Mortal Wkly Rep Kochanek KD, Murphy SL, Xu J, Arias E. Mortality in the United States. NCHS Data Brief 2017;293. US Department of Health and Human Services. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: 2012. Rosenman, Stephen J. "Preventing suicide: what will work and what will not.? The of Australia 169.2 [1998): 100-102. American Association. Diagnostic and Statistical Manual of Mental Disorders American Association; Arlington, VA. 2013. Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: a technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 2017. Simon, Thomas R., et al. "Characteristics of impulsive suicide attempts and attempters." Suicide and th?e~ Threatening Behavior 32.51 {2002): 49-59. Schiff, Lara B., et al. "Acute and chronic risk preceding suicidal crises among middle-aged men without known mental health andior substance abuse problems: An exploratory mixed?methods analysis.? Crisis: The Journal of Crisis intervention and Suicide Prevention 36.5 (2015): 304. World Health Drga nization. Risks to Mental Health: An Overview of Vulnerability and Risk Factors. August 2012. Caine, Eric D., et al. "Comprehensive, integrated approaches to suicide prevention: practical guidance.? injury prevention [2017): injuryprev?2017. iepub ahead of print) Cavanagh, J.T.D., Carson, A.J., Sharpe, M., 8; Lawrie, S.M. autopsy studies of suicide: A systematic review. Medicine, 2003; 33, 395405. Tables and Figures [attachments] Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 1780/1800 Rev 3.19.18 Draft 2.2 Short title: Vital Signs: Increasing Trends in State Suicide Rates and Contributing Circumstances Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Structured abstract [2511250 words) Background: Overall suicide rates have been rising in the United States. Examining state-level trends in suicide and its multiple contributing circumstances, can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates, by state and sex, among people aged 210 years, were assessed using data from the National Vital Statistics System. Changes in rates were examined across six consecutive three?year periods from 1999-2016. The National Violent Death Reporting System {2015), covering 27 states, was used to examine the precipitating circumstances among suicide decedents with and without known mental health problems (MHP). Results: Forty?four states saw statistically significant suicide rate increases from 1999?2016. In 25 states, rates increased by 30% or more. Male suicide rates increased significantly in 34 states while female rates increased significantly in 43 states. People with and without known MHP had both differing and similar circumstances precipitating suicide. Several circumstance, such as any relationship problems/loss [39.6 and 45.1%, ,o 5 any life stressorsfloss (49.7 and 54.2%, 5 .01), and recent crises (26.0 and 32.9%, 5 .01), respectively, were more likely among those without known MH P, but were common across groups. Conclusions: Suicide rates rose significantly across most states from 1999-2016. Varied circumstances beyond MHP alone contributed to suicides among people with and without known MH P. Implications for Public Health Practice: States can use a comprehensive public health approach based on the best available evidence to prevent suicide risks before they occur, identify and support people already at risk, prevent-reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {250f250 words) In 2016, nearly 45,000 suicides (15.6f100,000) occurred in the United States (US), among people 210 years Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 101th leading cause of death and is among the only leading causes to be increasing Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, increased more than 40% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strotegyfor Suicide Prevention(NSSP) calls for a public health approach to suicide with prevention efforts spanning across multiple levels individual, family/relationship, community, and societal of the social ecology. Such an approach underscores that suicide is rarely caused by any single factor alone, but rather, is multi?determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems (MHP) Other associated risk factors include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problemssolving skills, and prior suicide attempts, among others Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state?specific Flev 3.19.18 Draft 2.2 40 trends in suicide rates, assesses the multiple factors associated with suicide, and provides recommendations for 4 multi-Ievel comprehensive suicide prevention. 4: METHODS (ass/250 words} 43 Suicide rate estimates and trend analvses exclude data for persons <10 vears old because intent for self-harm 44 tvpicallv is not attributed to voung children. Age-specific suicide counts were tabulated based on National Vital 45 Statistics System coded death certificate records [internationoi Ciossificotion of Diseases 10"1 Revision 46 underlying?cause?of death codes X60?ii34, W810, Age?specific population estimates were obtained from 47 0.5. Census Bureaquational Center for Health Statistics bridged?race population data releases. 48 49 National and state?level suicide rate estimates were calculated for six consecutive three?veer aggregate periods 50 from 1999-2015. Rate estimates were age?adjusted to the 0.5. year 2000 standard population and expressed per 5] 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?veer data 52 aggregates, emploving weighted least squares regression with inverse-variance weighting. Modeled rate trends 53 are reported in terms of average annual percentage changes 54 55 Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting i i 56 5vstem were used to compare the characteristics among suicide decedents with and without known 5? current mental health problems (MHP). MHP are defined in as disorders and listed in the 5' 58 Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition except alcohol and other 59 substance Use disorders (captured aggregates data from three primary data sources: death 45 60 certi?cates, coronerfmedical examiner reports 6 with and without known MHP were compared using Chi?square tests; logistic regression was used to estimate 62 adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and raceiethnicitv. as RESULTS [596!600 words) '64 The most recent overall suicide rates (representing 2014-2015] ranged from 6.9 {District of Columbia] to 29.2 65 {Montana} per 100,000 persons per vear, a four-fold difference (Table Across the entire studv period, rates as increased in all but one state {Nevada}, with absolute increases ranging from +0.2 (Delaware) to +8.1(Wvomingl a? per 100,000. Percentage increases in rates ranged from +53% (Delaware) to 67.6% {North Dakota), with 63 percentage increases of at least 30% observed in half of all stata. 69 T0 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall {Table 1] 'il By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. 72 Nationallv, the model?estimated AAPC for the overall suicide rate was Bv sex, the national AAPC was "i3 +11% for males and +16% for females. i4 7'5 Suicide decedents with and without known MHP were compared. While both groups were To predominatelv male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of 7? being male [95% confidence interval and signi?cantlv greater odds of being racial/ethnic T8 minorities (odds ratio range: CI range - They also had significantlv greater 79 odds of perpetrating homicide-suicide (adjusted odds ratio 19.95% CI of firearm suicide laOR 80 1.5, 95% CI and of positive toxicology results for alcohol {5100 1.2, 95% CI Fifteen til percent ofthose with known MHP and 20% without ever served in the U.5. militarv. 32 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more 83 than those without MHP {19.8% vs. most freq uentlv by over?the?counter/otherwise unclassified drugs 34 opioids antidepressants or benzodiazepines Comment Izaf??l: Realized this wasn?t correct as stated. Did you guys notice I added this and is that how vou'd {Katie} want to say it? Fiev 3.19.18 Draft 2.2 85 All suicide decedents with known MHP and approximately 85% without had precipitating as circumstances information. People with MHF were more likely to have any substance abuse problems (31.6% vs. 8? 25%, While two-thirds ofthose with known MHP had a history of MH or substance abuse treatment 88 just oVer half were in current mental health treatment at the time of their deaths. Decedents without known MHP versus those with had greater 39 9O ikelihoo of wrelationship problemg?er?Ioss (1145.1 and 39.6%, 5 .01, aDFi 1.3, 95% CI specifically .3 9 intimate partner" Er'?ifi?'?i; 1363333321? iIiTs'a?ifci? 92 13.5%, and recently perpetrating interpersonal violence [3.0 and 1.4% 93 They also had significantly greater odds of am other life stressors, such as 94 criminal legal problems and 6.2% .01} or evictionfloss of home laQR 95 9%9?14?1?43 and and they were also more likely to have had a crisis within the preceding or as upcoming two weeks (32,9 Among both groups, the most 9? common crises were intimate partner and physical health problems. 93 Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but when a 99 release was indicated, they were more likely to he recently released from a correctional facility 4.5, 95% l?ii CI 3.2-6.3} or hospital laDR 1.3, 95% CI Among decedents with known MHP recently released from 10] an institution 42.3% were released from facilities. l02 Suicide decedents without known MHP leaveiag a suicide Iii} note [35.1 and 31.8% while decedents with known MHP, compared to those ID4 without MHP, more often had a history of suicidal ideation vs. 23.0%, pg .01] and attempts {29.4% vs. 10.3%. lii? Conclusions and Comments words] Ili'ir From 1999-2015, :14 states saw significant rate increases. Half of the states esperienced increases of 30% or ma more. Rates increased significantly in 34 states among males and 43 states among females. More research into Iii"? the causes of these trends is necessary 9 . IIU One important factor associated with suicide is MHP. Nearly half of suicide decedents in had a known I I I MHF. This group was challenged by comorbid substance abuse problems and histories of suicidal l2 ideation and attempts While two-thirds of people with MHP had a history of MH andfor l 3 substance abuse treatment and over half were currently in treatment at the time oftheir deaths, additional 114 support could help address the needs of this vulnerable population. This includes broader implementation of I I5 affordable and evidence-based treatments, such as doctor-patient collaborative care models and cognitive- llo behavioral therapy. Additionally. greater access to behavioral health providers, especially in Underserved areas is important, as is healthcare systems change that supports suicide prevention and patient safety through care I IS transitions ?9 While MHP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known 2ii I'leF. This group suffered more relationship problems and life stressors such as criminal?legal matters, IZI 122 suicides and attempts mom with minimal deliberation time, particularly among people without mental health IE3 disorders and who faced impending life crises [11, 12}. People with known MHP also experienced other life 24 stressors such asjob andfor financial problems, relationship problems, and physical health problems. These [25 findings point to the need to both prevent the conditions associated with mental health problems in the first Comment IZEWI: I took Tom?s suggestion and added the percentages instead of aORs. Not sure if this will be confusing now that some paragraphs are written one way [96} and others are ,written another [Bits]. Comment In?ll: This is a systematic review showing that socioeconomic factors are at least as important as MHP in suicide. Could instead link to something on the economy or the IBM report from 2002 which describes the many risk factors and the need to addreSs eviction/loss of home, and recent or impending crises, This is notevvorthy in light of ?ndings that suggest many Lthem. 1 126 1.27r 128 129 130 131 132 133 134 135 136 .137 138 139 140 141 142 143 144 145 146 147 148 149 150 .151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 Rev 3.19.18 Draft 2.2 place and the need to support people with to decrease their vulnerability to poor social, health, and economic outcomes These results underscore the importance of comprehensive state suicide prevention activities that go beyond a focus on MH treatment alone. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support], teaching coping and problemvsolving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help?seeking and positive social norms}, supporting family and friends after a suicide has taken place, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are underway planning and implementing a comprehensive approach to suicide prevention The study findings have at least three limitations. Related to state trend analyses, four states (MD, MA, RI, UT), rankings might have been impacted by large proportions of injury deaths of undetermined intent, or by decreased percentages of such deaths overtime, which likely include some unrecognized suicides. Second, related to circumstances, is not yet nationally representative. The most recent data from inclusive of 27 states, represents half of the U.S. population. Third, abstractors are limited to data from investigative reports. health information is not captured directly from medical records but from key informants family) via coroner/medical examiner reports. Therefore, completeness and accuracy of information is reliant an informant knowledge. Studies including in-depth interviews with key informants often see greater attributions to MH and substance abuse disorders It is likely that some people without known MHP in the current study were experiencing mental health challenges at the time of death that were either not known or reported by informants, or were not captured in primary data sources. Suicide is a growing public health problem. Mental illness is an important risk factor for suicide, and is one of many requiring preventive action. Data from and resources such as Preventing Suicide: a Technicai Package of Policies, Programs, and Practices [10] can help states and communities better understand their suicide problem and prioritize comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, 77?0-488-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 3Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC Refe re nces: 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 Rev 3.19.18 Draft 2.2 10. 11. 12. 13. 14. 15. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atlanta, GA: National Center for injury Prevention and Control. Retrieved March 15, 2018. 2016. Ivev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. Kegler, 5.8., O.M. Stone, and K.M. Holland, Trends in suicide by level of urbanization?United States, 1999?2015. MMWR. Morbidity and mortality weekly report, 2017. 66(10): 270. Kochanek, K., et al., Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. Rosenman, 5.1., Preventing suicide: what will work and what will not. The Medical Journal of Australia, 1998. 169(2): p. 100-102. Torguson, K. and A. O?Brien, Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. 2013': Washington, D.C. American Association, Diagnostic and statistical manual of mental disorders (058458). 2013: American Pub. Li, 2., et al., Attributable risk of and socia?ecanomic factors for suicide from individual?level, population?based studies: A systematic review. Social Science 8: Medicine, 2011. 72(4): p. 608616. Stone, D.M., et al., Preventing suicide: A technical package ofpoiicies, programs, and practice. 2017. Simon, T.R., et al., Characteristics of impulsive suicide attempts and attempters. Suicide and Life? Threatening Behavior, 2002. 32(51): p. 49?59. Schiff, L.B., et al., Acute and chronic risk preceding suicidal crises among middle?aged men without known mental health and/or substance abuse problems: An exploratory mixed~methods analysis. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 2015. 36(5): p. 304. World Health Organization, Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO, 2012. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practical guidance. Injuryr prevention, 2017: p. injurvprev-2017?042366. Cavanagh, J.T., et aI., autopsy studies of suicide: a systematic review. medicine, 2003. 33(3): p. 395?405. Tables and Figures (attachments) Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 180671800 Fiev 3.12.13 Draft 2 Structured abstract words) 2 Background: Overall suicide rates have been rising in the United States Examining state-level trends in 3 suicide and the multiple contributing circumstances can inform comprehensive state suicide prevention 4 planning. 5 Methods: Trends in age-adjusted suicide rates, by state and sex, among people aged all) years, were assessed (2. using data from the National Vital Statistics System. Changes in rates and state rankings were examined across 'i six consecutive three-year periods [lass-20m, zoos?20m, zoos-zoor, zoos-amp, 201142013. 2014-2015}. The ii National Violent Death Reporting System {2015}, covering states, was used to compare the precipitating 9 circumstances between suicide decedents with and without known mental health problems I'llI Results: Forty-four states saw statistically signi?cant suicide rate increases, over the period. In 30 states, rates ll increased by 25% or more. Male suicide rates increased in 34 states while female rates increased in 43. People 2 with and without known MHP had both unique and similar precipitating circumstances associated IS with their suicides, and many factors, such as relationship problems (39.6 and life stressors and 14 and recent crises [26.0 and respectively, were common to both. IS Conclusions: Suicide rates rose significantly across most states from 1999?2016. People with and without known In faced multiple circumstances contributing to their suicides. l7 Implications for Public Health Practice: To reverse upward suicide trends, states can use a multj-level public I?d health approach based on the best available evidence to: prevent multiple suicide risks before they occurs, identify and support people already at risk, prevent-rea?empts, and help after a suicide occurs. 20 INTRODUCTION 2 BACKGROUND AND PURPOSE {244(250 words] TOTAL 22 In 2016, nearly suicides occurred in the United States ll.l.S.i.1 Between 1999 and 2-316, 23 suicide rates increased among males and females, across racial/ethnic groups, and across urbanization lays-sis}4 24 Suicide, is the 10'? leading cause of death and is among the only leading causes to be increasing. Additionally, 25 rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, increased more 26 than 40% between mm and 2D15.5 Together, suicides and selfvharm injuries cost the nation more thanESES' 2? billion direct medical and work loss costs.1 Need to update i i i 28 The National-Strategy for Suicide Prevention calls tone-pu?b?cnh?ea?lth approach to suicide with prevention efforts 29 reaching across the social ecology individual, familyfrelationship, community, and societal levels]. Such an 30 approach highlights that suicide is rarely caused by any single factor but rather is mold-determined. 5 Despite 3 this, suicide prevention efforts primarily focus on identifying and treating mental illness le.g., depression}. 32 However, other associated risk factors exist and include social and economic problems, access to lethal means 33 log. substances, firearms} among people at risk, poor coping and problem?solving skills, and prior suicide 34 attempts, among others.is Expanded awareness of the multiple circumstances that contribute to suicide risk 35 apart from mental health problems alone, can help achieve substantial reductions in suicide rates. To assist 36 states in this goal and toward comprehensive suicide prevention planning, this study analyzes state?specific 3? trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention 38 recommendations. 39 METHODS words} Rev 3.12.18 Draft 2 Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Ciassrficatian of Diseases 10?? Revision underlying-cause-of death codes X60-X84, Y87.0, Age-specific population estimates were obtained from US. Census Bu reau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods from 1999-2016. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Data from 2015 from the 27' states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics among suicide decedents with and without known current mental health problems (MHP). MHP are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured separately). aggregates data from three primary data sources: death certificates, coronerirnedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and raceiethnicity. RESULTS (580,1600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rate trends indicated significant increases for 44 states, as well as for the US. overall (Table By sex, modeled rate trends indicated significant increases in 34 states for males and 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known MHP were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval and significantly greater odds of being racialfethnic minorities (odds ratio range: 1.0-2.1; 95% CI range They also had significantly greater odds of perpetrating homicide?suicide (adjusted odds ratio [308] 2.9, 95% CI of firearm suicide (30R 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the US. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.8% vs. most frequently by over-the-counterfotherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without had precipitating circumstances information. Two-thirds of decedents with known MHP had a history of MH or substance abuse treatment and were more likely to have any substance abuse problems {31.6% vs. 25%, .01). 2 Rev 3.12.18 Draft 2 86 Decedents without knovvn MHP had signi?cantly greater odds of any type of relationship problem laOR 1.3, 87 95% CI intimate partner problemskag? 1.4, 95% Cl laDR 88 1.4, 95% CI and recently perpetrating 89 had significantly greater odds of other life stressors, such as criminal legal problems laOR 95% CI 1.519] 90 or eviction/loss of home laDR 1.4, 95% Cl 1.24.5], and they were also more likely to have a crisis within the 9 preceding or upcoming two weeks laClFl 1.4, 95% CI Among both groups, the most common crises 92 were intimate partner and physical health problems. 93 Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but when a 94 release was indicated, they were more likely to be released from a correctional facility laOFl 4,5, 95% CI 3.2- 95 6.3] or hospital 1.3, 95% Cl 1.1v1.7].Among decedents with known MHP who Were recently released 96 from an institution 42.8% were released from facilities. 97? Suicide decedents without known MHP had significantly greater odds of leaving a suicide note laOl-l 1.2, 95% 98 Cl while decedents with known MHP more often had a history of suicidal ideation {40.8% vs. 23.0%, 99 5 .01] and attempts [29.4% vs. 10.3%, 5 IOU Conclusions and Comments [634}?00 words] lill From 19932015, 44 states saw signi?cant rate increases. One state [Nevada] saw a significant decline. Suicide 102 rates increased by more than 25% in 30 states and upwards of 57% in one {North Dakota] state. Rates increased 03 in 34 states among males and increased in 43 states among females. These increases may signal a narrowing of HM the suicide gender gap, historically weighted heavily towards males.11 More research into this troubling trend is [05 needed. [Up One important factor associated with suicide is MHP. Nearly half of suicide decedents in had a known It}? MHP. This group was challenged by comorbid substance abuse problems and histories of suicide IDS ideation and attempts While two?thirds of people with MHP had a history of MH andfor lill-i? substance abuse treatment and over halfwere currently in treatment at the time of their deaths, additional Ill support could help address the needs of this vulnerable population. This includes broader implementation of affordable and evidence-based treatments, such as doctor-patient collaborative care models and cognitive- 12 behavioral therapy.12 Additionally, greater access to behavioral health providers, especially in underserved areas I I3 is important, as is healthcare systems change that supports suicide prevention and patient safety through care ?4 transitions. 13 HS While MHP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known I I6 MHP. This groUp suffered moreiife stressors, especially related to relationships intimate partner problems, ii arguments or con?icts, recent perpetration of intimate partner violence], but also related to other life stressors ills such as criminal-legal matters, evictioniloss of home, and recent or impending crises (often related to the I I9 abovementioned known MHP also experienced life and other stressors apart from their [20 MHP. This group was more likely to experience fob andfor ?nancial problems. Also common were intimate l2] partner problems, physical health problems, and recent or impending crises] Lreferences on this. I22 These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH [23 treatment alone. Prevention strategies may include: strengthening economic supports leg. housing stabilization IE4 policies, household financial support], teaching coping and problem-solving skills, especially early in life to 25 manage everyday stressors and prevent future relationship problems; and promoting social connectedness to 12s increase a sense of belongingness and access to informational, tangible, emotional, and social support, as l2? needed. Dther strategies indicated by these results include creating protective environments reducing 3 Comment Can you say this? Wouldn?t saying this imply that if IPP is checked as a precipitating factor, then relationship problems also has to be checked? Is that how NUDRS coding of works? I thought these two categories were mutually exclusive in that could be checked without relationship problems being checked. (at .p This could be easily changed by just Lremoving "specifically" Comment vid5 = Wondering if instead of recapping results, we could comment on the profound and not often discussed impact that life stressors, especially those that rise to the level of a crisis, can have. The research that shovvs the time amount of deliberation {which is often very low] may also be helpful to cite here, as these types of can represent short?term problems lvs. longer term kproblems such as chronic mental illness] Comment tgs9 : I think this is worth considering. We have the space to include a concise point about how situational factors can contribute to immediate risk and that some suicides are impulsive. This could probably be included at the end of the paragraph since it applies to both Lthose with and without MH problem. Comment vid5]: Similarly, wondering if we could condense this instead of reiterating some of these results and instead comment on the concept of "functional impairment," as a reason why people with mental health problems might have more jobffinancial problems leg, they often have more inconsistent work histories as a result of trying to cope with their illness, perhaps have medical bills to deal with regarding treatment, periods of disability, etc. i. This IWould go a little deeper and interpret the results a bit. I?m sure we could find some Jk. ?ha 1 IZS [30 l3] 32 I33 134 135 136 [33 139 I40 14] I42 I43 I44 I45 I46 [47'r I48 I49 150 152 153 154 I55 156 15? LES I59 I00 l6 I62 I63 I64 [65 [66 lo? I68 169' I70 l?i'l I72 Rev 3.12.13 Draft 2 access to lethal means among people at risk, creating organizational and workplace policies to promote help- seeking and positive social norms), supporting people after a suicide has taken place to prevent survivors' risk, and assuring safe reporting by the media in order to prevent suicide contagion.11 These findings have at least three limitations. In four states (MD, UT, MA, rankings might have been impacted by large proportions of deaths of undetermined intent, or by decreased percentages of undetermined deaths over time, which likely reflect some unrecognized suicides. Second, is not yet nationally representative. This study used the most current data available and includes 2? states that represent half of the U.S.isopulatiori Third, abstractors of data are limited to data included in investigative 5 reports. For example, medical and MH information are not captured directly from medical records but from t6? informants family, friends} via coronerfmedical examiner reports. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and health problems based on described in interviews with family and friends. It is likely that some people without known mental health problems in the current study were experiencing mental health challenges at the time of death, but the lack of awareness about their mental health problems underscores the importance of addressing the range of contributing circumstances. 13 Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many. Resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices? and data from the can help states and communities prioritize comprehensive suicide prevention. References BE 1 CDC. Web-based Injury Statistics Query and Atlanta, GA: US Department of Health and Human Services, 2016. 2 World Health Organization. Preventing Suicide: A Global ImperatiVE. Geneva, Switzerland: World Health Organization; 2014. 3 Nev-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001?2015. MMWR Surveill Summ 4 Kegier SR, Stone OM, Holland KM. Trends in Suicide by Level of Urbanization United States, 199942015. MMWR Morb Mortal Wkly Rep 5 National Action Alliance for Suicide Prevention, Office of the Surgeon General. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, US Department of Health and Human Services, Of?ce of the Surgeon General; 2012. 6 Davidson, L, Potter, L., and Ross, lii. Surgeon lGeneral's Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 1999. 7 American Association. Diagnostic and Statistical Manual of Mental Disorders American Association,- Arlington, VA. 2013. Mercado MC, Holland K, Leemis RW, Stone OM, Wang J. Trends in emergency department visits for nonfatal self-inflicted injuries among youth aged 10 to 24 years in the united states, 2001?2015. JAMA Si Canetto, 5.5., and Sakinofsky. l. ?The gender paradox in suicide." Suicide and ?fe?threatening behavior 23(1): 1993: 1-23. 10 Stone OM, Holland ltM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: a technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, CDC: 2017. Comment Itgs?ii: This was in the 2015 55. I think it would be good to include it here too. It will be good to Lcon?rm the accuracy with Katie. Comment [tgaS]: I agree with the comment below that including the 90% estimate here could be problematic. It might be better to describe the alternative types of studies. I suggested an edit so you can see what i mean but this will need to be updated to be Lcol-isistent with what you cite. 173 174 175 176 177 178 179 180 .181 182 183 184 185 186 187 188 189 Rev 3.12.18 Draft 2 11 Isometsa E. autopsy studies?e review. European Acknowledgments Conflict of Interest None Corresponding author Stone Tables and Figures To address "While two?thirds of people with MHP had a history of mental health andlor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed." The highlighted point is only in the table right now- could add to results text. "People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job and/or financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead}. Rev 3.12.18 Draft 2 Structured abstract [2501'250 words) 2 Background: Overall suicide rates have been rising in the United States Examining state-level trends in 3 suicide and the multiple contributing circumstances can inform comprehensive state suicide prevention 4 planning. 5 Methods: Trends in age?adjusted suicide rates, by state and sex, among people aged :10 years, were assessed 6 using data from the National Vital Statistics System. Changes in rates and?ssase?Fa-ald-ngs?were examined across 7" Six consecutive three-year periods from 8 2914?2916}. The National Violent Death Reporting System [2015], covering 2? states, was used to compare the 9 precipitating circumstances between suicide decedents with and without known mental health problems 10 Results: Forty-four states saw statistically significant suicide rate increases; over the period. In 30 states, rates ll increased by 25% or more. Male suicide rates increased in 34 states while female rates increased in 43. People l2 with and without ?5436: known MHP had both differing and similar precipitating circumstances associated with their suicides, and many factors, such as relationship problems {39.6 and life stressors and [4 and recent crises [26.0 and rospedivelv, Were more likely among those with no known MHP, but l5 were common among both groupsConclusions: Suicide rates rose significantly across most states 1999-2016. A variety of circumstances 20 beyond MHP contributed to suicides among Ppeople with and without known MHoweced?mul-tiple 21 22 Implications for Public Health Practice: can use a multl-level public 23 health approach based on the best available evidence to: prevent multiple suicide risks before they occurs, 24 identify and support people already at risk, preventureattempts, and help after a suicide occurs. 25 INTRODUCTION 26 BACKGROUND AND PU RPDSE {244; 250 words} TOTAL 1300 2? In 2016, nearly 45,000 suicides {13.4f100,000l occurred in the United States Between 1999 and 2016, 28 suicide rates increased among males and females, across racialg?ethnic groups, and across urbanization levels.? 29 Suicide, is the 10?h leading cause of death and is among the only leading causes to be increasing. Additionally, 30 rates of Emergency Department visits for nonfatal self-harm injury. a key risk factor for suicide, increased more 3 than 40% between 2001 and 2015.5 Together, suicides and self-harm injuries cost the nation more thanEE-B 1500: Need to update 32 billion direct medical and work loss costs.1 33 The National Strategy for Suicide Prevention calls for a public health approach to suicide with prevention efforts 34 reaching across the social ecology individual, familylrelationship. community, and societal levels]. Such an I 35 approach highlights that suicide is rarely caused by any single factor but rather is Despite so this, suicide prevention efforts primarily focus on identifying and treating mental illness leg, depression}. 3? However, other associated risk factors exist and include social and economic problems, access to lethal means 38 leg. substances, firearms} among people at risk, poor coping and problem-solving skills, and prior suicide 39 attempts, among others.E Expanded awareness of the multiple circumstances that contribute to suicide risk 40 apart from mental health problems alone, can help achieve substantial reductions in suicide rates. To assist 1 Rev 3.12.18 Draft 2 states in this goal and toward comprehensive suicide prevention planning, this study analyzes state-specific trends in statesuicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METHODS (2457250 words) Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Classification of Diseases 10?? Revision underlying-cause-of death codes NED-X84, Y87.0, Age-specific population estimates were obtained from U.5. Census Bu reau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods from 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and earpressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCS). Data from 2015 from the 22 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics among suicide decedents with and without known current mental health problems (MHP). MHP are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured separately). aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racefethnicity. RESULTS (5801'600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 5.9 (District of to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to +5.16% (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall (Table 1). By sex, modeled rate trends indicated significant increases in 34 states for males and 43 states for females. Nationally, the model-estimated AAPC for overall suicide rates was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known MHP were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval and significantly greater odds of being racial/ethnic minorities {odds ratio range: 102.1; 95% Cl range They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known lleP and 20% without ever served in the US. military. Ftev 3.12.18 Draft 2 85 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more rComment Need tothinkabout 86 than those without lleF {19.8% vs. most frequently by over-the?counterfotherwise unclassi?ed drugs what to do with this sentence. Perhaps delete and say "More research into these 8? opioids antidepressants or benzodiazepines _trends is needed.? Comment [14:15]: Wondering if instead of recapping results, we could comment on the profound and not often discussed impact that life stressors, especially those i that rise to the level of a crisis, can have. i The research that shows the time amoum of deliberation (which Is often very low} i may also be helpful to cite here, as these i types of can represent 4' I i i 88 All suicide decedents with known MHP and approximately 35% without iN=9,357l had precipitating 89 circumstances information. Two?thirds of decedents with known MHP had a history of MH or substance abuse 3 9i) treatment [612%] and were more likely to have any substance abuse problems [31.6% vs. 25%, i i 9 Decedents without known MHP had signi?cantly greater odds of any type of relationship problem 1.3, 92 55% CI specifically intimate partner problems 1.4, 95% Cl arguments/conflicts iaDR 93 1.4, 95% CI and recently perpetrating interpersonal violence 2.0, 95% CI They also 94 had signi?cantly greater odds of other life stressors, such as criminal legal problems 1.7, 95% CI 151.9} 95 or evictionfloss of home 1.4, 95% CI and they were also more likely to have a crisis within the as preceding or upcoming two weeks 1.4, 95% CI Among both groups, the most common crises 97 were intimate partner and physical health problems. short-term problems (vs. longer term ,?problems such as chronic mental illnessig Comment I think this is worth considering. We have the space to include a concise point about how situational factors can contribute to immediate risk and that some suicides are impulsive. This . . . . . . 98 Decedents virithout known MHP had signi?cantly lower odds of recent release from any institution, but when a 5 could probably be included release was indicated, they were more likely to be released from a correctional facility 4.5, 95% CI 3.2- IUO 6.3) or hospital 1.3, 95% Cl Among decedents with known MHP who Were recently released the paragraph since it applies to both Lthose with and without MH problem. IUI from an institution 5.213% were released from facilities. ?12 Suicide decedents without known had significantly greater odds of leaving a suicide note iaOFl 1.2, 535% Cl 1.14.2}, while decedents with known l'v'iHF' more often had a history of suicidal ideation {40.8% vs. 23.0%, [04 5 .01] and attempts [29.4% vs. 10.3%, Jo 5.01). [05 Conclusions and Comments words] [06 From 1999-2016, 44 states saw significant rate increases. ,l rates increased by more than 25% in 30 states and upwards of 57% in one [North Dakota) state. Rates increased [(18 in 34 states among males and increased in 43 states among increases may signal a narrowing of i 109 the suicide gender gap, historically weighted heavily towards n'iales.11 ]lVlore research into this troubling trend is If.) needed. One important factor associated with suicide is MHP. Nearly half of suicide decedents in had a known ?2 MHP. This group was challenged by comorbid substance abuse problems and histories of suicide I I3 ideation and attempts While two?thirds of people with MHP had a history of MH andfor I I4 substance abuse treatment and over half were currently in treatment at the time of their deaths, additional I IS support could help address the needs of this vulnerable population. This includes broader implementation of [6 affordable and evidence-based treatments, such as doctor-patient collaborative care models and cognitive- behavioral therapy.?- Additionally, greater access to behavioral health providers, especially in underserved areas I I8 is important, as is healthcare systems change that supports suicide prevention and patient safety through care I lit transitions. 1? 120 While l'leP are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known MHP. This grouo suffered morelife stressors, especially related to relationships intimate partner problems, [22 arguments or conflicts, recent perpetration of intimate partner violence}, but also related to other life stressors 23 such as criminal?legal matters, evictionl'loss of home, and recent or impending crises (often related to the ,l 24 abovementioned experienced life and other stressors apart from 125 126 12? 128 129 13D 131 132 133 134 135 I36 137? 133 139 Mi) I4I I42 I43 I44 I45 I46 I4r I48 I49 150 151 152 154 155 156 157 158 159 160 lo] 162 163 164 165 166 16? Rev 3.12.18 Draft 2 MHP. This group was more likely to experience Eob andforflnancial problems. Also common were intimate partner problems, physical health problems, and recent or impending crises]? These results point to the need for comprehensive suicide prevention that goes beyond a focus on I'v?lH treatment alone. Prevention strategies may include: strengthening economic supports le.g g. housing stabilization policies. household financial support}, teaching coping and problem?solving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Dther strategies indicated by these results include creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help- seeking and positive social norms}, supporting people after a suicide has taken place to prevent survivors? risk, and assuring safe reporting by the media in order to prevent suicide contagion.12 These findings have at least three limitations. In four states UT, MA, rankings might have been impacted by large proportions of deaths of undetermined intent, or by decreased percentages of undetermined deaths over time, which likely re?ect some unrecognized suicides. Second, is not yet nationally representative. This study used the most current data available and includes 27 states that represent half {49. of the U. S. bopulatiori" Thir_d, abstractors of data are limited to data included in investigative reports. For example medical and 1le information are not captured directly from medical records but from key informants leg, family, friends} via coroneri'medical examiner reports. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy health problems based on described in interviews with family and without known mental health problems in the current study were experiencing mental health challenges at the time of death, but the potential lack of awareness about their mental health problems underscores the importance of addressing the range of contributing circumstances. 13 Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many. Resources such as CDC's Preventing Suicide: Technical Package of Policies, Programs, and Practices13 and data from the can help states and communities prioritize comprehensive suicide prevention. References BE 1 CDC. Web-based Injury Statistics Query and Reporting System Atlanta, GA: US Department of Health and Human Services, 2016. 2 World Health Organization. Preventing Suicide: A Global Imperative. Geneva, Switzerland: World Health Organization; 2014. 3 Ivey-Stephenson AZ, Crosby AE, Jack SPD, Halieyesus T. Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death - United States, 2001?2015. MMWR Surveill Summ 201?;66l18l??16. 4 Kegler SR, Stone DM, Holland KM. Trends in Suicide by Level of Urbanization United States, 1999?2015. MMWR Morb Mortal Wkly Rep 5 National Action Alliance for Suicide Pretrention, Office of the Surgeon General. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: US Department of Health and Human Services, Office of the Surgeon General; 2012. IS Davidson, L., Potter, L., and Ross, lii. Surgeon General?s Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 1999. ?1 Comment Similarly, wondering if we could condense this instead of reiterating some of these results and instead comment on the concept of "functional impairment,? as a reason why people with mental health problems might have morejobi'financial problems le.g., they often have more inconsistent work histories as a result of trying to cope with their illness, perhaps have medical bills to deal with regarding treatment, periods of disability, etc.l. This would go a little deeper and interpret the results a bit. I?m sure We could ?nd some _references on this. Comment This was in the 2015 55. I think it would be good to include it here too. It will be good to Lcon?rm the accuracy with Katie. Comment tgs9 : lagree with the comment below that including the 90% estimate here could be problematic. It might be better to describe the alternative types of studies. I suggested an edit so you can see what i mean but this will need to be updated to be ,consistent with what you cite. JII. JR. 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191. 192 193 194 Rev 3.12.18 Draft 2 7 American Association. Diagnostic and Statistical Manual of Mental Disorders American Association; Arlington, VA. 2013. 8 Mercado MC, Holland K, Leemis RW, Stone DM, Wang]. Trends in emergency department visits for nonfatal self-inflicted injuries among youth aged 10 to 24 years in the united states, 2001-2015. JAMA 9 Ca netto, 5.5., and Sakinofsky. l. "The gender paradox in suicide.? Suicide and life-threatening behavior 28{1): 1998: 1-23. 10 Stone DM, Holland KM, Bartholow BN, Crosby AE, Jack SPD, Wilkins N. Preventing suicide: 3 technical package of policies, programs and practices. Atlanta, GA: US Department of Health and Human Services, 2017. ll lsometsa E. autopsy studies?a review. European Acknowledgments Conflict of Interest None Corresponding author :3 Stone Tables and Figures To address "While two-thirds of people with MHP had a history of mental health andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed." The highlighted point is only in the table right now? could add to results text. ?People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead}. Rev 3.6.13 Draft Really the more important ordering? my Background: rates have been rising in the United States Examining state-level trends in, and I: had contributing circumstances to, suicide, can inform comprehensive state suicide prevention planning. r? Methods: Trends in age-adjusted suicide rates, by se-x?and?state Eta-rte and seaE among people aged 3:10 years, fr; Were evaluated using data from the National Vital Statistics System. Changes assessed across six consecutive three?year periods [1999-2001, 2002-2004, 2008-2010, 2011-2013, 2014-2015}. Data from 2015 from the National Violent Death Reporting System, across 2? stat?, were analyzed to compare precipitating circumstances between suicide decedents with and without known mental health problems. Results: Statistically significant upward rate trends were identified for 44 states. For the U.5. overall and for 30 states individually, empirical rates increased by at least 25% over the study period. People with and without known mental health problems experienced a range of circumstances contributing to their suicides. including recent crises and problems related to substance use, relationships, jobffinancial issues, and criminal?legal matters. Conclusions: Suicide rates have risen significantly in the U.S.'and across most states from 1999-2015. No single factor alone contributes to suicide. Differing circumstances contribute to suicides among those with and without mental health problems. Implications for Public Health Practice: To reverse upward trends in suicide, states can use a population?based public health approach inclusive of evidence-based strategies across multiple levels (individual, familyfrelationship. community, societal], focused on preventing risk before it starts, identifying and supporting people already at risk, preventing?reattempts, and caring for survivors after a suicide. INTRODU CTIDN BACKGROUND AND PURPOSE {2101' 250 words] TOTAL In 2016, nearly 45,000 suicides {lint/100,000) occurred in the United States lLl.S.i.? While overall rates have been declining globally,?1 rates of suicide in the US increased between 1999 and 2016, among males and females, across racial/ethnic groups, and across urbanization levels.?1 Emergency department visits for nonfatal self?harm injuries increased more than 40% between 2001 and 2015.5 In 2015, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work less costs.1 Suicide is rarely caused by one factor; rather. the risks are often numerous and occur at multiple levels?individual, familylrelatlonship, community, and societal? Despite this, prevention primarily centers on mental illness, leg, depression, bipolar disorder}. Other factors associated with suicide include social isolation, economic downturns, access to lethal means substances, firearms) among people at risk, childhood adversity, lack of coping and problem-solving skills, loss of a friend or family member to suicide, a prior suicide attempt, and unsafe media portrayals, among others? While the Surgeon General called for a comprehensive public health approach to suicide prevention in 1999, to date, most states struggle to make this a reality} To better assist states, this study analyzes trends in state suicide rates, assesses the multiple factors associated with suicide, and provides prevention recommendations. METHODS 1259;250 words} Suicide rate estimates and trend analyses exclude data for persons <10 years old. Age?specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {loternotionoi Classification of 1 Rev 3.6.13 Draft 1 Diseases 10?1 Revision underlying-cause-of death codes X60-XB4, YBIO, Age-specific population estimates were obtained from U.S. Census Bureau National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods covering years 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Data from 2015 from the 27 states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics, including precipitating circumstances, of deaths by suicide among decedents with and without known current mental health problems (M HP). Mental health problems are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence (captured separately). aggregates data from three primary data sources: death certificates, coronen?medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and racefethnicity. RESULTS (6961'600 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia (0.0.1) to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table 1). Across the entire study period, rates increased in all but one state (Nevada), with increases ranging from +0.2 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to (North Dakota), with percentage increases of at least 25% observed in over half of all states (30), as well as nationally. Modeled suicide rate trends significant increases for 44 states, as well as for the U.S. overall (Table 1). By sex, modeled rate trends significant increases in 34 states for males and in 43 states for females. Nationally, the model?estimated AAPC for overall suicide rates was By sex, the national AAPC was +11% for males and for females. Suicide decedents with and without known mental health problems (MH P) were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP had 2.3 greater odds of being male (95% confidence interval (CI) 222.5), and significantly greater odds of being racial/ethnic minorities (odds ratio range: 1.02.1; 95% Cl range They also had significantly greater odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide (30R 1.6, 95% CI 1.54.7), and of positive toxicology results for alcohol 1.2, 95% Cl Fifteen percent of those with known MHP and 20% without ever served in the US. military. Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP (19.8% vs. most frequently by over-the-counteriotherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 85% without had known precipitating circumstances. Decedents without known MHP had signi?cantly greater odds of any type of relationship problem 1.3, 95% CI specifically intimate partner problems 1.4, 95% CI argumentsiconflicts 1.4, 95% CI and recently perpetrating interpersonal violence Rev 3.6.18 Draft 1 2.0, 95% CI Two-thirds of decedents with known MHP had a history of MH or substance abuse treatment (62.2% and were more likely to have any substance abuse problems (31.6% vs. 25%, 5 .01), Suicide decedents without known MHP had significantly greater odds of other life stressors, such as a criminal legal problems 1.195% CI or eviction/'Ioss of home 1.4, 95% They had significantly lower odds of recent release from any institution, but when a release was indicated, they were significantly more likely to be released from a correctional facility 4.5, 95% CI or hospital 1.3, 95% CI Among decedents with known MHP who were recently released from an institution, 42.8% were from institutions. Those with known MHP also more frequently had job andfor financial problems (16.8% vs. 15.6%; ,o 5 Decedents without known MHP had significantly greater odds of a recentfimpending crisis 1.4, 95% CI When the type of crisis was known, it was most frequently related to an intimate partner physical health criminal legal issues family relationships or a job Over one fourth of decedents with a known MHP also had recent or impending crises, most frequently related to an intimate partner physical health or family relationships Decedents without known MHP had significantly greater odds ofcriminal legal crises 1.6, 95% CI and significantly lowered odds ofjob-related crises 0.7, 95% Cl 050.8). Suicide decedents without known MHP had significantly greater odds of leaving a suicide note 1.2, 95% CI while decedents with known MHP more often had a history of suicidal ideation (40.8% vs. 23.0%, ,o 5 .01) and attempts (29.4% vs. 10.3%. .0 .01) Conclusions and Comments (7351700 words) From 1999-2016, age?adjusted suicide rates among people a 10 yea rs increased 25.4% overall. Forty?four states saw significant rate increases and one (Nevada) state saw a significant decline. Suicide rates increased by more than 25% in 30 states and upwards of 50% in some. Among females, rates increased in 43 states and rates among males increased in 34. This signal of increasing vulnerability of females towards suicide aligns with recent reports that identified a 63% increase in middle?aged female suicide rates between 1999-2014 and an annual increase of 18.8% per year in emergency department visits for self-inflicted injuries among young females, aged of 10 and 14, in the period 2009-2015110 These increases may hint at a narrowing of the suicide gender gap, historically weighted towards males lay?a?Fat-ieref?dl? More research into this troubling trend is needed. One important factor associated with suicide is mental health problems. Nearly half of people in this study had a known mental health problem. This group was challenged by comorbid substance abuse problems and histories of suicide ideation and attempts While two?thirds of people with MHP had a history of mental health andfor substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed. This includes the need for broader implementation of affordable and evidence-based treatments, such as doctor?patient collaborative care models and cognitive?behavioral thera py.12 Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is healthcare systems change that supports suicide prevention and patient safety. 12 While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MHP. This group suffered more life stressors, especially related to relationships intimate partner problems, arguments or conflicts, recent perpetration of intimate partner violence), but also related to other 3 Rev 3.6.13 Draft 1 life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises {often related to the abovementioned factors}. People with known MHP also experienced life and other stressors apart from their MH P. This group was more likely to experience job and/'or financial problems. Also common were intimate partner problems physical health problems and recent or impending crises These results point to the need for comprehensive suicide prevention that goes beyond a focus on MH treatment alone. These strategies may include: strengthening economic supports housing stabilization policies, household financial support}, teaching coping and problem-solving skills and other pro-social norms, especially early in life to manage everyday stressors and to prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments reducing access to lethal means among people at risk, and creating organizational and workplace policies to promote help-seeking and positive social norms), and supporting people after a suicide has taken place to prevent survivors? risk and to assure safe reporting by the media in order to prevent suicide contagion.12 These findings have at least three limitations. In four states, Maryland (MD), Utah Massachusetts and Rhode Island state rankings might have been impacted by large proportions of deaths of undetermined intent (MD), which often represent cases where a suicide determination wasjudged not to be conclusive, or by decreased percentages of undetermined deaths over time (UT, MA, Second, is not yet nationally representative. Currently, 40 states, the District of Columbia, and Puerto Rico participate in but the most recent available data year includes 27 states as others joined the system later. Third, abstractors of data are limited to data included in investigative reports. For example, medical and mental health information are not captured directly from medical records but from informants family, friends) via coronerfmedical examiner reports. Therefore, knowledge of the informant impacts completeness and accuracy of the information reported. This may explain some of the discrepancy between the prevalence of mental health disorders reported here and the 90% statistic frequently cited in other studies. 13 Suicide is a growing public health problem and mental illness is an important risk factor for suicide, but is just one of many associated factors. Resources such as Preventing Suicide: a Technicai Package of Policies, Programs, and Practiceslzand the National Violent Death Reporting System can help states and communities prioritize comprehensive suicide prevention. References BE UPDATED) 1 Wisaars fatal injury reports 2 World report 3 NCHS Data brief 2014 4 Kegler et al 201? 5 WISQARS Nonfatal injury reports 6 National Strategy for Suicide Prevention 7 Davidson, L., Potter, L., and Ross, V. [1999} Surgeon General?s Call to Action to Prevent Suicide. Public Health Service Rockville, MD. 8 9 10 Melissa?s paper 11 Need ref 12 Technical package Acknowledgments Conflict of Interest None Corresponding author Stone Rev 3.6.13 Draft 1 Tables and Figures Options for ref 11 - :Hawton K, van Heeringen K. Suicide. Lancet. 2009;373: erasioelsosz-x -. Canetto SS, Sakinofsky I. The Gender Paradox in Suicide. Suicide 8t Life-Threatening Behavior. 1998; 28:1-23. From Katie: I went through the latest draft w/Alex?s comments, comparing the results and discussion. Most lined up, although there were a few places where I noticed some mismatch t? is from the draft, bold are my comments]: ?From 1999-2016, age-adjusted suicide rates among people 2 10 years increased 25.4% overall.? We give the national (overall) number only in the abstract, not in the results. ?Suicide rates increased by more than 25% in 30 states and upwards of 50% in some?. We only mention one [North Dakota} that is over 50% in the results {rather than "some?}. "While two-thirds of people with MHP had a history of mental health and/or substance abuse treatment and over half were currently in treatment at the time of their deaths, much more support for this vulnerable population is needed.? The highlighted point is only in the table right now- could add to results text. "While MHP are a significant contributor to suicide, 54% of suicide decedents in the current study did not have a known MHP. This group suffered more life stressors, especially related to relationships intimate partner problems, arguments or conflicts, recent perpetration of intimate partner Didn't notice this the first time around, but the highlighted portion should read "recent perpetration of interpersonal violence" - that variable is not specific to IPV. ?People with known MHP also experienced life and other stressors apart from their MHP. This group was more likely to experience job andfor financial problems. Also common were intimate partner problems physical health problems and recent or impending crises The numbers in the highlighted portion are in the table, but not specifically cited in the results (because we report odds ratios instead). Morbidity and Mortality Weekly Report Morbidity and Mortality Weekly Report Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Journal: 1 Morbidity and Mortaiity Weekly Report Manuscript Type: Manuscript CDC author Manuscript ID lcoc-zola-onz I Date Submitted by the Author: 25-Apr-2018 Complete List of Authors: Stone, Deborah; National Center for Injury Prevention and Control, Division of Iviolence Prevention Simon, Thomas; CDC, NCIPC Fowler, Katherine; Centers for Disease Control and Prevention, Division of Violence Prevention Kegler, Scott; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control Yuan, Kerning; National Center for Injury Prevention and Control, Division of Violence Prevention Holland, Kristin; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control Nev-Stephenson, Asha; Centers for Disease Control and Prevention, DVP Crosby, Alexander; National Center for Injury Prevention and Control Jurisdiction - Country [select all that apply): United States Multistate >15 states: I Yes Jurisdiction States and US. Territories (if 15 states, select all that apply}: Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files movies) online. DStone_Figure 1 .emf Page i of 36 Morbidity and Mortality Weekly Report l-D 4.25.18 Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Sci];1 Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, l?y'lDl Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among people aged 1:10 yea rs, by state and sex, across six consecutive three-year periods {1999-2016}, were assessed using data from the National Vital Statistics System for 50 states and Washington, DC. Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 19992016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors {54.2% vs and recentfimpending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with but were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides {15.6f100,000 occurred in the United States (US), among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels (2, Suicide is the 10Th leading cause of death and is one ofjust three leading causes that are increasing (1, 4). Additionally, rates of emergency department visits for nonfatal selhharm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self?harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strategyfor Suicide Prevention{NSSP) (5) calls for a public health approach to suicide prevention with efforts Spanning across multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather, is multi- determined. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal people, providing treatment for mental health problems and preventing re-attempts In addition to MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms} among people at risk, and poor coping and problemsolving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, 1 nuscriptcenl ra comfmmun i-D Morbidity and Mortality Weekly Report Page 2 of 36 4.25.18 can help reach the nation?s goal of reducing suicide rates 20% by 2025 (7). To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides options for multi-level comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for people aged 210 years only, as determining suicidal intent in younger children can be difficult (3). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Ciassification of Diseases 101". Revision, underlying?cause-of death codes KEG-X84, Y87.0, 003). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridgedvrace population data releases. National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronen?medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racei?ethnicity. RESULTS The most recent overall suicide rates {representing 2014-2016) varied four-fold, from 6.9 (D.C.) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +5.53% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (Table 1, Figure 1). Modeled suicide rate trends indicated significant increases for 44 states, for males (34 states) and females (43 states}, as well as for the U.S. overall (Table 1). Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 2? states. While all decedents were predominately male (Table 2; 76.8%) and non?Hispanic white those without known MHP, relative to those with MH P, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5) and racial/ethnic minorities (DR range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicidersuicide (adjusted odds ratio 2.9, 95% CI 2.13.8). 2 nuscriptcenl ra cornimmvin Page 3 of 36 Morbidity and Mortality Weekly Report l-D 4.25.18 Among adult decedents, 20.1% of those without known MHP and 15.3% of those with MPH ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.5%] and less likely to die by (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% such as opioids 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP (N29,407j and approximately 85% without known MHP had available circumstances information [Table People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without, versus those with, known MH P, had significantly greater likelihood of any relationship problem/loss (45.1% vs. specifically intimate partner problems (30.2% vs. (12.5% vs. and recently perpetrating interpersonal violence vs. They also were more likely to have experienced any life stressors (54.2% vs such as criminal-legal problems (10.7% vs. or eviction/loss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems were the most common types and did not differ by group. Similarly, among people without versus with MHP, physical health problems (23.2% and 21.4%] and job/financial problems {15.6% and 16.8%) were commonly experienced by both groups. Decedents without known MHP had significantly lower odds of recent release from any institution (a0R=0.5, 95% but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. hospital (43.7% vs. or other facility alcoholfsubstance treatment) 95% than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%] and prior suicide attempts (10.3% vs. More than one in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- female suicide rates between 1999-2014 (3). Additional research into the specific causes of these trends is necessary. Fortunately, data from the 27 states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. 3 nuscriptcenl ra l-D Morbidity and Mortality Weekly Report Page 4 of 36 4.25.18 Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention is often oriented towards identification of suicidal people, treatment of MHP and prevention of reattempts. Additional focus on non-mental health factors, further upstream, is essential to a public health approach as the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship problems and other life stressors such as criminal?legal matters, evictionfloss of home, and recent or impending crises. Similarly, people with MHP often experienced relationship problems and other life stressors such as job/financial and/or physical health problems. These findings point to the need to both help people manage the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two?thirds of this group had a history of any mental health andfor substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physicalfmental health problems] Other strategies include creating protective environments leg, reducing access to lethal means among people at risk, creating organizational and workplace policies to promote helpvseeking, easing transitions into and out of work for people with MHP and other life challenges}, supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, UT) might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders (13), however many methodological variations across studies exist (14). It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known MHP suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices (12] to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. 4 nuscriptcenl ra Page 5 of 36 l-D Morbidity and Mortality Weekly Report 4.25.18 Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No con?icts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 270488-3942 Author Af?liations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Centerfor Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow?Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed], Hindman], Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev-2017-042356. 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo 0. Critical issues in autopsy studies. Suicide Life Threat Behav 2006;36l5l1491-510. Word Count: 1904! 1800 Impsih?mcma nuscriptceril ra comfrrimvvi mummAwM?a kD Morbidity and Mortality Weekly Report Page I5 of 35 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons (Change tram Prior Period} Current Overall Overall State 5211 State Rate Change 2:32"; 1999 2991 2992 2994 2995 - 2997 2995 - 2919 2911 2913 2914 2915 Flank (State Flank) 1 (State 2 5591 12.311112} 12.7 1+ 9.4} 12- 91+ 9.21 1351+ 9.9} 14.51+ 9.51 1541+ 9.9} 1.5 95 152.911 1122 3.1 15221 25.4 9511115} US. Male 20.9 {1115} 2121+ 0.4} 21 31+ 0. 01 22.5 (+1.31 2351+ 1.0} 24.5 1+ 1.0} 1.1 113-4011 Female 11115} 5.0 1+ 0.3} 3-1-01 1+ 0.4} 6.2 1+ 0.5) 8.3 1+ 2.5 ?15 113-4011 Both 14.3 1n2a1 13.41- 9. 91 14.1 1+ 9. 51 15.51+1.5} 1541+ 9.71 17.51+1.1} 5 95 11:12. 951 25 3.1 131} 21.9 951331 AL Male 25.1 1122} 23.41.71 24. 4 1+27.51+1.11 29.1 1+ 1.5} +1.3 951112 951 Female 5.1 111251 4. 519.31 591+ 9.21 .11+1. 11 541+ 9.31 7.9 1+ 9.7} 2. 5 95 1112. 911 Both 21.9 1n2e1 24.5 1+ 3.1.2} 25.41- 9.51 25.5 1+ 3.4} 1. 7 95 1112. 951 2 7.51 4} 32.4 95 1131 AK Male 332195} 35.1 1+ 4.9} 35. 9 1+9. 51 49.1 1+1.2} 49.1 1- 9.11 42.91+2.5} 1. 4951 1.112 911 Female 5.5 {nfa} 11.4 1+ 2.91 9.51-1.51 11.1 1+ 1.2} 9.91- 1.21 13.2 1+ 3.4} 1.7 95 n15 Beth 17.5 1n2a1 15.5 1+ 9.2} 19.1 1+ 9.5) 19.1 1- 9.91 29. 4 1+ 1. 31 29.9 1+ 9.5} 1.9 95 1112.91) 15 3.1 132} 12.3 95 142} A2 Male 29.3 1n2e1 39.2 1+ 1.9} 39.5 1+ 9.41 39.2 1- 9.51 32. 9 1+ .91 32.4 1+ 9.4} 9.5 95 {1:12.951 Female 2.1 111121 7.5 1+ 9.4} 5.2 1+ 9.71 5.5 1+ 9.5121+ 9. 51 9.9 1+ 9.51 2.2 95 1132.91) Beth 15.5 1n2e1 1551+ 9.3} 1521+ 9.51 17.5 1+ 1.2.9} 2.2 95 1112.91} 12 5.7114} 35.5 95115} AH Male 25.71n2a1 25.7 1+ 9.9} 2721+ 9.51 25.21+1.9} 31713. 51 33.5 1+ 1.9} 1.5 95 1112.95) Female 5.5 {ma} 5.9 1+ 9.3} 5.2 1+ 9.41 7.9 1+ 1715-9141 9.5 1+ 2.1} 3.5 95 {112.91} 55111 19.51n1a1 1131+ 9.71 11.91- 9.31 12.91+1.9} 11. 519 .11 12.1 1+ 9.3} 9.9 951112.951 45 1.5145} 14.5 95145} CA Male 17.911151 1541+ 9.5} 17.71- 9.71 19.1 1+ 1.41 15. 919 .21 1921+ 9.31 9.5 95 n25 Female 4.1 111251 5.9 1+ 9.9} 4.91- 9.1} 5.4 1+ 9.5} .31-9 .11 5.5 1+ 9.3} 1.7 95 1112.95} Both 17.31n2e1 19.21+1.9} 19.91- 9.21 29.9 1+ 1.9} 2151+ 1.51 23.2 1+ 1.5} 1.5 95 11:12.91) 5 5.9112} 34.1 951221 CD Male 25.5 1n1a1 39.9 1+ 2.31 39.51- 9.41 31.5 1+ 1.9} 33. 41+ 1. 91 35.3 1+ 2.9) 1.4 95 1112.911 Female 7.9 91121 5. 2 1+ 1.9.91 19. 1 1+ 1. 91 19.4 1+ 9.3} 2.5 95 1112.91) Beth 951515199171 9 1 1+ 9 21 19.21+1.1} 11. 91+ 9.51 11.5 1+ 9.5} 1.5 95 1112.95} 45 1.9143} 19.2 951341 CT Male 15.41n2e1 14. 51-1 51 15 91+ 9.41 15.51+1.51 17. 51+ 1.91 17.31- 9.31 9.9 95 n15 Female 3.5 11151 3.5 1+ 9.9.7} 9+19.51 5.2 1+ 1.3} 3.5 95 1112.951 Page 7 of 36 Morbidity and Mortality Weekly Report 1 :1 T311011.I 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 4 Age-Adjusted Annual Flate per 100,000 Persons (Change from Prior Period) Current Overall Overall 5 State Sex ?10:13:: E: State Rate Change Eileen-f"; 1999 - 2001 2002 - 2004 2005 2007' 2008 2010 2011 - 2013 2014 - 2016 Flank 5 {State Rank} 1 (State Hgnk) 3 Beth 13.61n1a) 12.21- 1.4) 11 3 1- 0 3) 13.61+ 1.7) 14.2 1+ 0.6) 1441+ 0.2) 0.3 33 r175 42 0.3 150) 5.3 313150) 9 DE Male 23.0 (ma) 20.3 1- 2.22.7 1- 0.4) 23.5 1+ 0.3) 0.6 ?13 r115 10 Female 5.3 1n7a) 5.01- 0.2) 601641+ 1.5) 6.21- 0.2) 1.6 34 1175 1; Both 5.31n7a) 6.4 1+ 0.5) 4010.7) 6.3 1+ 0.3) 0.3 31. n15 51 +1.0 143) 16.1 ?79145) 13 DC Male 10.71n7a) 11.1 1+ 0.4) 10.3 1? 0.3) 1271+ 2.4) 10.0 1- 2.6) 11.7 1+ 1.7) 0.3 47.. r175 14 Female 1 7 We) *1 231+ 0.6) T) 3.3 1+ 1.0) 2.6 1? 0.7) 3.6 1+ 1.0) 3?1 0. 3) 3.5 07.. r115 15 Both 14.3 (Ma) 15.2 1+ 0.4) 14.3 1? 0.3) 16.3 1+ 1.4) 16.3 1- 0.0) 16. 4 1+ 0.1) 0.3 I34.16205) 23 +1.6 145) 10.6 34143) 16 FL Male 24.31n7a) 24.4 1+ 0.1) 23.6 1? 0.3) 2621+ 2.6) 25.6 1? 0.6) 25. 6 1- 0.1) 0.5 n15 1; Female 6 31n7a) 6.3 1+ 0.5) 6.3 1+ 0.0.5) 31+ 0. 3) 1.4 47.. (112.01) 19 Both 12.3 (n13) 1321+ 0.3) 12.31? 0.3) 13. 2 1+ 0. 3) 1371+ 0.5) 15. 0 1+ 1 .3) 0.3 ?in r115 33 2.1 140) 16.2 144) 20 GA Male 22.1 (ma) 23.1 1+ 1.0) 21.3 1? 1.3) 21.3 1+ 0. 6) 22.6 1+ 0.7) 2441+ 1.7) 0.5 34.. We 21 Female 5.01n7a) 4.31- 0.2) 4.6 1- 0.2) 51+ 0. 3) 5.3 1+ 0.3) 61+ 0. 3) 2.1 34 1p<.05) 2?2 Both 12.3 We) 11.1 1- 1.3) 10.31? 0.7) 14. 51+ 4. 1) 14.41? 0.1) 1521+ 0. 3) 2.0 34;. r175 35 2.4135) 13.3 133) 3: HI Male 20.41n1a) 17.21- 3.1) 15.3 1? 1.3) 2131+ 6. 7) 22.5 1+ 0.5) 24. 311+ 1 .3) 2.1 34 r115 25 Female 5.4 1n7a) 5.0 1- 0.4) 5.5 1+ 0.5) 7.1 1+ 1 5.) 6.21- 0.1175 26 Both 17.31n7a) 13.2 1+ 2.0) 13.3 1- 0.3) 2161+ 3. 3) 21.3 1+ 0.3) 24.7 1+2 .3) 2.3 33 (p201) 6 7.51 6) 43.2 ?16 1 7) 27 10 Male 23.41n1a) 33.1 1+ 4.7) 31.1 1? 2.0) 34. 3 1+2 3 3) 341.6 073113405) 23 Female 72 (nia) 6.1 1- 1.1) 6.1 1+ 0..1+3 2.3) 4.4 %1p<.05) :3 Both 3.31n7a) 3.31? 0.1) 3.7 1- 0.1) 10. 6 1+2 0 3) 111.0) +1.5 %1p<.05) 44 2.3 133) 22.3 132) 31 IL Male 17.1 1n7a) 16.71- 0.4) 16.21? 0.4) 17.6 1+ 1.4) 13. 511+ 0. 3) 13.31+1.3) +1.1 34 32 Female 3 71n7a) 3.61- 0.0) 3.3 1+ 0.2) 21+ 0. 4) 4. 511+ 0. 4) 21+ 0. 6) 2.4 07.113231) 33 Both 13.0 (nfa) 13.7 1+ 0.7) 1441+ 0.7) 1431+ 0. 5) 16.41+1 4) 17.1 1+ 0. 7) +1.3 %1p<.01) 26 4.1 123) 31.3 33125) 34 IN Maia 22.4 We) 23.2 1+ 0.3) 24.41+1.2) 241.6) 1.5 34;. (pc?l) :2 Female 4 61n1a) 5.0 1+ 0.4) 5.3 1+ 0.6.61? 0.2) 2.7 %1p<.01 MD Morbidity and Mortality Weekly Report Page 8 of 35 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:13:59? State Rate Change 1999 - 2991 2992 - 2994 2995 2997 2999 - 2919 2911 - 2913 2914 - 2919 Flank 5 (State Flank) ?7 (51919 33,1111 1* Beth 11.511119) 13.21+1.41 12.51- 041 14.21+1.41 +2.1 31 +4.31201 +35.2?191151 IA Maia 20.5 111191 22.1 1+ 1. 51 20.5 1- 1.41 23 3 1+ 2. 51 2501+ 2. ?1 25.? 1 +1.5 ?19 1132.051 Female 3 ?111191 5111+ 1. 01 5.3 1+ 0.5151+0.3.5 ?19 1p<.011 Both 13.311119) 15.1 1+ .1 1551+ 0.?1 15. 31?0 5.1 1+ 2. 41 19.41+1.51 2.2 191119.011 15 5.01111 45.0 ?191 51 KS Maia 22.? 111191 25. 01+ 2. 31 2551+ 1.51 25. 511.51 1.9 ?19 1112.011 Female 4.5 1n191.01+ 1.415.? 1 0.5.1 3.2 ?19 1p<.051 Both 14.1 111191 15.41+1.31 15.? 1+ 1.31 15 2151 1521+ 2.01 19.31 1 11 +1.5 191119011 20 5.21151 35.5 ?191151 KY Maia 25.0 111191 2551+ 1.91 2531+ 1.41 2721?15} 30.1 1+ 2. 91 31.?1 1+ 1 .51 1.4 191119.011 Female 4.5 19191 5.2 1+ 0.41 5 1 1+ 0.51 5.1 1+ 0.11 1+ 0511+ 0. 51 3.2 ?19 1112.01) Beth 13.1 111191 12.91? 0.21 134 1+ 0.41 1351+ 0.31 14 41+ 0. 31 1?.0 12 .51 +1.5 ?19 1132051 2? 3.5 1271 25.3 ?191251 LA Maia 22.9 111191 22.31? 0.51 2241+ 0.11 23.3 1+ 0.5) 23 1+ 0. 51 2?.31+ 3.51 1.1 ?19 919 Female 4.5 111191 1? 0.11 5 21+ 0.51 4.9 1? 0.21 5.1 1+ 1.21 1.41 2.5 ?19 {1:19.051 Both 145111191 13.51? 0.51 14 41+ 0 51 1541+ 1.01 15 51+ 3. 51 15.51? 0.41 2.2 ?19 1112.051 21 4.0 1251 2?.4 ?191291 ME Maia 25.0 111191 22.91? 2.11 2451+ 1.?1 25.? 1+ 1.11 31 1 1+ 5. 41 29.5 1? 1.31 +1.5 ?19 1p<.051 Female 5.31n191 5.31? 0.01 5.21? 0.11 501+ 0.?1 1.51 0.31 3.1 ?19 1p<.051 Both 10.011119) 10.3 1+ 0.3) 10.1 1? 0.21 1021+ 0.11 10 1+ 0. 51 1051+ 0.11 0.5 ?191p<.051 4? 0.5 149 5.5 ?19 143 ?511 MD Maia 17.511119) 1?.51+ 0.11 1?.31? 0.51 1?.71+ 0.41 15. 2 1+ 0. 51 15.01? 0.21 0.2 ?19 1119 Female 3.5 111191 3.5 1+ 0.41 3.5 1+ 0.01 1? 0.21 4.1 1+ 0.41 4.5 1+ 0.41 1.3 ?19 1p<.051 Both 4 19191 1+ 0.1.01 5.5 1+ 0.4) 1001+ 0.31 2.3 ?19 1119.01) 45 2.5 134 ?1 35.3 ?19 120 ?1 MA Male 12.1 111191 1251+ 0.?1 13 31+ 0 51 1541+ 2.11 15.21? 02} 1501+ 0.51 2.0 191119.011 Female 3.3111191 2.51? 0.41 4 01+ 1.01 3.51? 0.11 451+ 1.01 4.51? 0.21 3.0 ?19 1.92.051 Both 11.5 111191 1251+ 0.?1 1251+ 0.41 1351+ 1.01 1451+ 0.?1 1551+ 1.11 +1.9 ?19 1132.011 33 3.9 1251 32.5 ?19 {241 Ml Male 20.0 111191 2051+ 0.51 2151+ 0.?1 2251+ 1.31 2351+ 1.01 2501+ 1.21 1.5 ?19 199.011 Female 4.4 19191 4.5 1+ 0.41 5 0 1+ 0.21 5.5 1+ 0.51 5.9 1+ 0.31 1+ 0.51 2.5 ?19 199.011 Page 9 of 36 Morbidity and Mortality Weekly Report 1 :3 T311100.I 1. Trends in Suicide Rates among Persons 2 10 Years of l?lge,r by State and Sex, National Vital Statistics System, 1999 2016 4 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall 5 State Sex ?10:35? State Rate Change 5 1333 - 2001 2002 - 2004 2005 - 2007 2003 - 2010 2011 - 2013 2014 - 2010 Hank 5 (State Flank) ?7 (State ank) a 3 Both 10.? (1113) 11.5100) 1241+ 0.8) 1201+ 0.5] 14.21+1.3) 1501+ 0.91 2.3 %1p<.01) 38 4.3113) 40.5 ?/51 81 9 MN M515 13.311175) 13 3 1+ 1.1) 2041+ 1.0) 2031+ 0.3) 2231+ 1.3) 2331+ 0.4) +1.7 c11, 1115.01) ?3 Female 3.3 11175) 21 0. 3) 4 31+ 0.3) 5.1 1+ 0.4) 5.3 1+ 0.3) 331+ 1.2) 4.2 31. 1115.01) 1; 35111 12.311175) 14 1 1+2) 14.71+ 0.3) 1551+ 0.3) 1531+ 0.1) 1521- 0.3) 1.1 351115.05) 33 2.3 133) 17.3 35140) 13 1713 Male 22.3 11175) 24. 3 1+ 1. 7) 25.1 1+ 0.3) 23.3 1+ 1.7) 25.31- 0.3) 25.3 1- 0.3) 0.7 35 1175 14 Female 4.3 1n75) .0107) 551+ 0.5) 5.5 1- 0.0) 3.4 1+ 0.3) 3.2 1- 0.2) 2.4 35 1115.01) 15 31101 14.711175) 14.1 1-0 3) 1541+ 1 3) 1301+ 0.7) 1731+ 1.7) 2001+ 2.3) 2.2 31. 1115.01) 13 5.3 115) 33.4 31.117) ?5 11110 1111515 25311175) 23.713) 2531+ 1 3) 2331+ 1.0) 2331+ 2.3) 3221+ 3.3) 1.3 351115.05) 1; Female 5.411175) 4+1 0.1) 31 1+ 0.7) 3.3 1+ 0.2) 741+ 1.1) 331+ 1.2) 3.2 33 1115.01) 19 35111 21.1 11175) 22. 3 1+ .4) 2331+ 1 0) 2471+ 1.1) 2371+ 2.0) 2321+ 2.5) 2.1 073 1114.01) 1 3.0 1 2) 33.0 071.111) 20 MT Male 33.3 11175) 7 3 1+ 0. 4) 3331+ 2.5) 33 7 1-0 41.0 1+ 1.4) 4551+ 4.4) 1.3 31. 1115.01) 21 Female 3 711175) 41+ 1. 3) 3.41- 0.1) 10 01+ 1 .3) 1231+ 2.3) 13.1 1+ 0.5) 4.3 o731115.01) 22 Beth 12.711175) 1221- 05) 1231+ 0.4) 11 .7-10 .3) 13.51+1.3) 14.31+1.3) +1.0 E15 1175 40 2.1 142) +132 375 143) 3: NE Male 22.2 11175) 20.71- 1.5) 20.3 1- 0.4) 13. 3 1-0 .5) 22.0 1+ 2.2) 2331+ 1.3) 0.3 073 1175 25 Female 3.3 1575) 4.2 1+ 0.4) 5.1 1+ 0.3) 0-11.1) 5.5 1+ 1.4) 531+ 0.3) 2.3 31. 575 23 35111 23.311175) 22.31- 0.3) 22.1 1 0.5) 22.13 0. 5) 2141-12) 23.1 1+ 1.3) - 0.2 31+ 1175 3 - 0.2151) - 1.0 073151) 27 NV Male 38.3 (1113) 35.1 1? 1.6) 35. 32.5 3.0) 3541+ 2.3) ?7?5 H73 23 Female 3.311175) 351+ 0.5) 331+ 0.1) 10 01+ 0. 41 1031+ 0.3) 1121+ 0.3) 1.5 35 1114.01) :3 311111 13.511175) 12.51 1.0) 1331+ 0.3) 15. 2 1+ 1 .3) 1531+ 0.3) 2001+ 4.2) 2.7 31. 1115.05) 17 3.5 1 3) 43.3 341 3) 31 NH Male 22.511175) 21.1 1- 1.4) 2171+ 0.3) 24. 31+3 .1) 2541+ 0.3) 3031+ 5.2) 2.2 351115.05) 32 Female 5.3 11175) 4.31? 05) 531+ 1.0) 2+1 0. 4) 3.3 1+ 0.4) 331+ 3.2) 3.3 31, 1112.05) 33 35111 7.31n75) 7.71- 0.1) 7.51- 0.2) .01+ 0. 5) 331+ 0.3) 321+ 0.4) +1.3 073 1115.05) 50 1.5 147) 13.2 375135) 34 NJ Male 13.011175) 13.1 1+ 0.0) 12.31- 0.5) 13.71+ 1.1) 1451+ 0.3) 1431+ 0.1) 0.3 =11. 1115.05) :2 Female 3.2 11175) 2.31? 0 3) 301+ 0.0) 3 1- 0.1) 3.3 1+ 0.3) 4 41+ 0.3) 2.3 073 1175 MD Morbidity and Mortality Weekly Report Page 10 of 36 Table 1. Trends in Suicide Rates among Persons 2 10 Years of A96, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex ?10:13:: :3 State Rate Change 1999 - 2991 2992 - 2994 2995 2997 2999 - 2919 2911 - 2913 2914 - 2919 Flank 5 (State Flank) (State 33,111) 1* Beth 22.0 (1179} 22.01? 0.1} 21.61? 0.2} 23.0 1+ 1.2} 24.1 1+ 1.1} 2601+ 1.3} 1.1 319 (pa.05} 4 4.0 (24} 16.3 31.133} NM M919 36.6 11179} 37.7 1+ 0.3} 36.4 1- 1.2} 35. 6 1?0 37.1 1+ 1.3} 40.7 0.4 1179 Female 6.511179} 7.4 1 6. 2 1+ 0.7} 10. 71+ 2. 6} 1171+ 0.3) 12. 0 1+ 0. 3} 3.3 ?79 (pa.05} Both 7.211179} 7.1 1? 0.1} 0.6} 351+ 1.1} .3?01 2.1 ?79 (pa.01} 43 2.1 141} 26.6 679127} NY Male 12.511179} 12.21? 0.3} 12. 3 1+ 0.7} 13 3 1+ 1 15.41+ 1.4} 14. 510 3} 1.4 ?7.3 (pa.05} Female 2.711179} 2.61? 0.1} 01+ 0.3} 51+ 0. 5} 421+ 0.7} 61+ 0 5} 4.2 ?79 1pa.01} Both 13.6 (979} 13.51? 0.1} 1371+ 0.1} 14 2 0. 5} 1451+ 0.4} 15. 3 1+ 0. 6} 0.6 ?79 (pa.01} 34 +1.7 (44} 12.7 0.4147} NC Male 22.7 (1179} 22.7 1+ 0.0} 22.21? 0.6} 23. 3 1+ 1.1} 2331+ 0.0} 23,311+ 0. 6} 0.4 ??79 1179 Female 5.6 (9790.6} .0?012} 67 1+ 0.7} 61+ 0. 3} 2.0 34.. (pa.05} Both 13.3 (1179} 14.6 1+ 1.3} 16.0 1.4} 16 610.6} 1641+ 1.3} 20. 3(+2 2.3 (pa.01} 14 7.6 1 5} 57.6 ?791 1} ND Male 21.4 (1179} 24.6 1+ 3.2} 2601+ 3.4} 27 1 1?0 23.6 1+ 2.5} 32 71+ 3. 0} 2.5 ?79 (pa.01} Female 5.6 (1179) 4.519 1.0} 3.7 1? 0.6} 6.7 1+ 1.0} 51+ 1 3.3 1179 Both 11.611179} 1231+ 0.6} 13.1 1+ 0.6} 13 4 1+ 0. 2} 1461+ 1.4} 1561+ 1.0} 2.0 (pa.01} 32 4.2121} 36.0 (13} OH Male 20.4 (1179} 2031+ 0.5} 22.2 1.3} 2 1 1?0 2421+ 2.1} 25. 5 1+ 1 1.5 (pa.01} Female 4.0 (979} 4.7 1+ 0.7} 3 1+ 0.1} 31+ 0. 5} 6.2 1+ 0.3} 0. 6} 3.4 ?79 (pa.01} Both 17.0 (1179} 16.51? 0.6} 17.2 0.6} 16. 4 1.1} 2071+ 2.3} 23 512.6} 2. 3 ??79 (pa 05} 7 6.4 (10} 37.6 679112} UK Male 26.5 (1179} 27.3 1? 1.2} 2761+ 0.5} 30 3 2. 5} 33.4 1+ 3.??79 (pa. 05} Female 6.6 (1179} 6.41? 0.2} 7.5 1+ 1.1} 10?015} 6.51+ 1.6} 10 31+ 1 2. 3 ?79 (pa. 05} Both 16.4 (979} 17 7113} 17.7 0.0} 16 0. 3} 1361+ 1.2} 21 1 1+ 1 +1.6 ?79 (pa 01} 13 4.6 (16} 26.2 34.126} Male 27.4 (1179} 23512.1} 26.51? 0.3} 23 .5 0} 3141+ 1.6} 33. 0 1+ 1 +11 (pa. 01} Female 6.5 (1179} 7.1 1+ 0.6} 7.7 1+ 0.6} 0. 7} 6.6 1+ 0.4} 61+ 0. 3} 2. 7 ?31 (pa 01} Both 12.1 (1179} 1251+ 0.4) 12.6 0.3} 13 3 1} 1501+ 1.1} 16. 3 1+ 1 2.0 (pa.01} 30 4.1 (22} 34.3 a79121} Male 21.0 (979} 21. 31+ 0.3} 2131+ 0.6} 231 2471+ 1.7} 26.1 1+ 1 +1.5 ?79 (pa.01} Female 4.2 (11790.0} 4+0( 601+ 0.6} 11+ 1. 1} 3.5 ?79 (pa.01} Page 11 of 36 MD Morbidity and Mortality Weekly Report Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:35? State Rate Change 1000 - 2001 2002 - 2004 2005 - 200? 2000 - 2010 2011 - 2013 2014 - 2010 Flank 5 (State Hank) ll (State Beth 9.411110) 9.01? 0.3) 9.010 0) 12.313 3) 11.91? 0.9) 12 3 1+ 0. 2.3 31: 1p<.05) 43 3.2130 34.1 33123 RI Male 15.41n1a) 1521? 02) 1431?0 .3) 21 21+ 3. 4) 19.21? 2.0) 19 3 1+ 0. 4) 2.2 34.. n15 Female 4.0 (Me) 3.3 1? 3.3 1+ 0.4) .11+ 1 .3) 5.1 1+ 0.0) .11+ 1 .0) ?11. 1p<.05) Both 12.31012) 1301+ 0.2) 13.? 1+ 14. 91+1.2) 13.01+1.1) 1+ 1 2.3 {pi-<01) 23 4.9 11?) 33.3 34.110) SC Male 21.3 111.12) 2251+ 1.2) 22.31? 0.1) 24. 312.2) 23.1 1+ 1.5) 3 01+ 1.9) 1.3 1p<.Dl) Female 5.4 1n1a) 1? 301+ 1.3) 21+ 0. 2) 1+ 0.3) .41+ 1 4.) 3.4 {pl-:05) Beth 15.?1nia) 1531+ 0.1) 1?.1 1+ 1 19.31 2.2) 19.? 1+ 0.4) 2231.9) 2.5 3911:1301) 10 ?.01 44.5 31:1 3) 30 Male 27.311110) 2331?13) 2?.9 41+ 1. 3) 30.1 1+ 2.2) 3201+ 1.9) 33 3 1+ 1 .3) 1.3 1p<.01) Female 4.211110) 531+ 1.3) 1+ 0. 3) 3.3 1+ 2.0) 1? 1.0) 11. 31+4 .0) 5.3 31; 1p<.01) Both 14.311112) 1521+ 0.3) 13.141+ 0.3) 1?.21+ 1.1) 1?.21+ 0.0) 13. 2 1+ 1.0) +1.4 ?13 1p<.D1) 22 3.5 123) 24.2 31.131) TN Male 25.1 (Ma) 25.4 1+ 0.3) 2331+ 1 3) 23. 0 1+ 1.2) 23.3 1+ 0.3) 29. 3 1+ 1 .2) 1.2 ?11. 1p<.01) Female 5 4 11112) 3.3 1+ 0.9) 1+ 0.4) 5+1 0.3) 3.91? 0.3) 31+ 0. 1.9 10-505) Both 12.21012) 12.? 1+ 0.3) 12.31? 0.4) 13. 21+ 0. 9) 1331+ 0.3) 14. 1+5 0. 9) 1.1 31.. 1.02.01) 41 2.3 13?) 13.9 133) TX Male 20.4 11112) 20.9 1+ 0.5) 20.4 1- 0.3) 22.0 1+ 1 .3) 22.2 1+ 0.3) 23.1 1+ 0. 9) 0.9 1p<.05) Female 4 3 (n12) 5.4 1+ 0.3) 5.0 1? 0.4) 201+ 2) 5.3 1+ 0.4) 41+ 0. 3) 1.3 ?11. Both 1? 2 (r112) 1901+ 1.3) 1321? 20. 21+ 2. 0) 2401+ 3.3) 2521+ 1. 2) 31010301) 5 3.0 1 3 W) 43.5 31.1 4 W) UT Male 23.21012) 31.1 1+ 2.9) 29.41?1.?) 32.1 1+2 3?.31+ 33. 0 1+ 0. 2) 2.1 ?19 1132.05) Female 3.31012) 0.3) 0.1) 1031+ 2.1) 1231.0) 4.4 ?20 113-501) Beth 13.2 We) 1321+ 3.0) 14.9 1? 1.3) 13 13.? 1+ 2.1) 19. 1+ 1 0) 2.4 ?11. 13 3.41 9) 43.3 34.1 2) VT Male 23.31n1a) 23. 3 1+ 4. 3) 24.31? 4.0) 3131.01+ 32 51+ 1.5) 1.9 31010205) Female 4.3 11112) 5.2 1+ 0.9) 3.4 1+ 1.3.3 1112.01) Beth 12.311112) 12. ?10.1) 1291+ 0.3) 13. 31+ 0. 1431+ 0.1132.01) 3? 2.2 139) 1?.4 34.141) VA Male 21.3 (Ma) 21. 31? 0.2) 21.01? 0.4) 22. 51+ 1. 5) 2331+ 1.2) 23.9102) 0.9 ?11. 1p<.05) Female 5.3 (Me) 521? 0.1) 591+ 301? . 3) 3.4 1+ 0.3) 91+ 0. 5) 1.3 34.. 1p<.05) 0054010145de N-D Morbidity and Mortality Weekly Report Page 12 of 35 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:13:: :1 State Rate Change 1999 - 2001 2002 - 2004 2005 - 2007 2008 - 2010 2011 - 2013 2014 - 2018 Flank 5 (State Flank) ll (State Rink) 4* Both 14-81n1a) 15.410 5) 14.81? 0.8) 15.?1+ 0. 9) 18 81+ 0.9) 1181+ 1.0) +1.1 %1p<.05) 24 2.8138) 18.8 31:13?) WA Male 24.?1n1a) 25. 21+0 .5) 24.1 1? 1.1) 25.1 28 01+ 0. 9) 2?.1 1+ 1.1) 0.8 Female 5.9 (nta) 41+ 0. 8) 8.2 1? 0.2) 9+10 71+ 0. 8) 851+ 0.8) 2.5 ?to (10:91) Both 15.81n1a) 1?21+ .8) 18.?1?05) 18?001.?) 19.21+3.2) 21.41+ 2.2) +1.13% n/s 11 +5.8113) 4131184114) WV Male 27.21n1a) 30 1 1+ 2.9) 2881?15) 27.810) 51+ 3.9) 3351+ 2.0) 1.1 ?In nis Female 5.3 (nta) 51+0 .1) 8+1 0.2.2) Both 13.1 (nta) 1351+ 0.4) 14.0.1+05) 1501+ 1.0) 1531+0.3) 1851+12) 1.5 ?to 1p<.01) 28 +3.4129) +258 ?31: 130) WI Male 21.?1n1a) 22. 21+ 0..2) 24.41+ 0. 4) 25.? 1+ 1 3) 1.1 til/5113401) Female 5.1 1n1a51+ 0.1) 7151+ 1.0) 2.5 34., 1p<.01) Both 20.?1n1a) 23.14 .7) 22. 51?0 9) 2541.8) 28. 91+ 3. 5) 28.81? 0.1) 2.3 ?15 1p<.Dl) 3 8.1 1 1) 39.0 ?rt: 1 9) W?r? Male 34.8 (Ma.2) 4?.1 1+ 5. 8) 44.81? 2.4) 1.8 ?to 1p<.05) Female (nta) 8.2 1+ 0.8) 9.2 1+ 0.9) .4+01.2) 10. 1+ 1. 4) 12.8 1+ 1.9) 3.2 ?to 1p<.01) Ftates are age-adjusted to the LLB. year 2000 standard. Model-estimated average annual percentage change based on all reporting periods: p-value indicates statistical significance of trend: nr?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 - 2018. Hanks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically significant difference. 'l Overall rate change is between the first (1999 - 2001) and last {2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2001) and last {2014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). ll Ftate based on 20 suicides. Differences between ranks do not necessarily imply a statistically significant difference. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Page 13 of 36 mumm-h-wM?s ND Morbidity and Mortality Weekly Report Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted {95% (n=20,446} prob em* Mental Health Square (95% c? a} {$9,407} Problem {n=11,039) Sex Male 15302063) 6,469l683} 9,233l83.6) p<.01 Female 4,744l232) 2338812} 1,306l16.4) p<.01 Age" 10?24 2,804l13.7) 1,211l12.9} p<.01 25?44 6,456l31.6) 3,036l323} p<.05 45-64 7271887.?) 3,820i40.6] 339885.31 p<.01 65+ 3,468l17.0] 1,340l14.2] 2,128l19.3] p<.01 Race/ethnicity White, non?Hispanic 17,102l83?) 8,165l86.8} 8,937l810) p<.01 Black, non-Hispanic 1,228l5.0] Mil-4.4) 81347.4} p<.01 American Indiaanlaska Native, non-Hispanic 378l1.8} 112(1.2) 2669.4) p<.01 Asian, non?Hispanic 576l2.8} 235(2.5) 3418.1) p<.05 Hispanic 463(4.9) 6335.7} p<.05 Other 66(03) 21(02) 4503.4) p<.05 Extended demographics Ever served in military? 3,429l17.8) p<.01 Homeless 240l1.2} 104(11) 136{1.3) Incident Type Single suicide 20,063l982) 9318(991] p<.01 Homicide followed by suicide 319l1.6} 64(07) 255{2.3} p<.01 Multiple suicides 64(03) 25(03) 39(0.4) Method Firearm 93091435) p<.01 5,907l28.9] 2,940i313} p<.01 Poisoning 1,861l19.8] p<.01 mumm-h-wM?u ?0 Substance class causing death?? Other over-the-cou nter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for 2 1 substance? Substance detected Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive Morbidity and Mortality Weekly Report 94431.4} 524(203} 219(73} 931304.41 4,442l40.6] 8,554l41.8) 499(53} 75151372) 73597} 55591311) 2,214t4os) 555(3ss) 508(32.7) 544(34s) 458(251) 195(105) 730(33) 5553005} 5,409i57.5} 4,253r453} 1,238i29.1} 1,6398%} 3,866i41.l} 215(55) 375(102) 719(22.7) 3,103l33.0} 355(31.1} 336(29.4} 155(137} 155(13.7) 24(2.1) 315{7.4} 4295355) 325(212} 283(53) 3319855] 360{9.2} 75103.1} 479(205) p<.01 p<.Dl p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 ones-0.7) 0.703.703) earns?1.1) cares-0.4) 0.30.1433) 0.703.507) Dalila?1.0) news?1.0) 0.503.507) Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan, Minnesota. New Hampshire. New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma. Oregon, Rhodc Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Page 14 of 36 Page 15 of 36 Morbidity and Mortality Weekly Report Decedent had been identi?ed as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. *5 Odds ratio refleCts the risk among those without known mental health problem relative to those with knewn MHP. 1' Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age. sex, race and ethnicity. Known MHP was used as the reference group. Decedents were aged [0 years and older, as per standard in the suicide prevention literatureDemmlnator is decedents aged 13 years of age and older reported military service status. Denominator is decedents who died by poisoning. including overdose. Denominator is with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. 43-h- new mumm-lh-wM?n ND Morbidity and Mortality Weekly Report Page 16 of 36 Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted on? Problem+ Mental Square [95% [95% cu Heakh Problem Suicide with known circumstances 18.764813) 9.407(100] p<.Dl Mental Health Am; Current Diagnosed Mental Health Problem? 7.076(752) Anxiety disorder Bipolar disorder Schizophrenia 50915.4} PTSD ADDKADHD 22642.4} Unknown 760l3.1} Current depressed mood 3,962l42.1) 3,076l32.9) p10 years with and without known mental health problems National 3 Violent Death Reporting System, 2? states.? 2015 4 No ltnown 5 Known mental mental health health problem?. problem. no. chi? Adjusted on1 5 Characteristics Total no. (?111) Square DR5 (95% Cl) (95% Cl) .7 Suicide with known cimumstances 16.?64 (91.6) 9.40? (100) 9.35? (64.6) p<0.01 3 Mental Health 9 Any Current Diagnosed Mental Health Problem? (7'52) 10 Anxiety disorder 1.519 (16.6) 11 Bipolar disorder 1.431 (15.2) 12 Schizophrenia 509 (5.4) PTSD 424 (4.5) 13 ADDJADHD 226 14 Unknown ?60 (6-1) - - - - 15 Current depressed mood 3.962 (42.1) 3.0?6 (32.9) p<0.01 Substance Problems 1 Any Current substance problem 5.319 (26.3) 2.9?6 (31.6) 2.343 (25.0) p<0.01 17' Alcohol problem 3.266 1.662 (19.6) 1.406 (15.0) p<001 16 Other substance problem 3.064 (16.4) 1.?63 (16.6) 1,316 (14.1) p<001 Treatment l9 Current mental healthisutlstance ahuse 5.141 (54.0) 64 p<0.01 0.01 (0.01?0.01) 0.01 (0.01?0.01) 20 treatment 21 Ever treated for mental healihisubstance (35.6) 6.323 394 (4.2) 0.02 (0.02+0.02) 0.02 (0.02~0.03) 22 problem Relationship ProblemsiLoss 23 Any relationship problemiloss 1.943 (42.4) 3.726 (39.6) 4.222 (45.1) p<0.01 1. 3 (1. 2 1.3) 1 3 24 Intimate partner problem 5.096 (27.2) 2.2?0 (24.1) 2.626 (30.2) ps0.1.4 (1.3-1.5) 25 Perpetrator ofinterpersonal violence in past 414 (2.2) 131 (1.4) 263 (3.0) ps0.01 2. 2 (1. 6? 2. 2.0 month 26 Victim ofinterpersonal violence in past month 64 (0.4) 53 (0.6) 31 (0.3) p<0.05 0.6 0.6 2? Family relationship problem 1.6?1 (6.9) 6?3 (9-3) ?96 (6.5) 0.9 (0.6-1.0) 1.0 (0.9-1.1) 23 Other relationship problem (nonqntimate) 403 (2.1) 202 (2.1) 201 (2.1) - 1.0 1.1 Argument or conflict (not speci?ed) 2.914 (15.5) 1.2?6 (13.6) 1.636 p<001 1.3 1.4 29 Death of a loved one (any) 1.49? (8.0) 626 (6.6) 6?1 ps0.01 0.6 0.9 30 Non?suicide death 1.161 (6.3) 64? (6.9) 534 p<0.01 0.6 0.9 31 Suicide of family or friend 3?9 (2.0) 21? (2.3) 162 p<0-01 0.6 32 Other Life Stressors Any life stressor 9.?43 (51.9) 4.6?5 5.066 (54.2) p<001 1.2 (1.1-1.3) 1.1 (1 1?1.2) 33 Recent criminal legal problem 1.536 (6.5) 566 (6.2) 1.002 p<0.Dtherlegal problem ?46 (4.0) 3?6 3?0 (4.Physical health problem 4.1?9 (22.3) 2,012 (21.4) 2.16? (23.2) p<0.JobiFinanciai problem"1 2941 (16.2) 1530 (16.6) 1411 (15.6) p<0.Eviction or loss of home ?22 (3.6) 31? (3-4) 405 (4.3) p<001 1.3 1 .4 (1. 2? 1-6) 3? School problem?? 162 (19.9) ?0 92 (21.9) 1. 3 (0 9?1.Recent release from an institution? 1.412 941 (10.2) 4?1 (5.1) p<0.01 0-5 (0. 4?0- 5) 0 5 (0. 4-0. 5) Jailiprisonidetention facility 203 (14.4) 82 121 p<001 3. 5 (2. 9) 4.5 39 Hospital 51? (36.6) 311 (33.0) 206 ps0.01 1.6 1.3 40 hospitaliinstitution 469 (33.2) 439 30 (6.4) p1:0.01 0.1 0.1 (0. 1?0. 1) 41 Other (includes alcoholiSA treatment facilities) 223 (15.6) 109 (11.6) 114 (24.2) ps0.Recent or Impending Crisis 4'2 Crisis within past or upc0ming 2 5.525 (29.4) 2.444 (26.0) 3.061 (32.9) ps0.01 1.4 1 4 (1.3-1.5) 43 Intimate partner problem crisis 1966 (35.6) 654 (34.9) 1114 (36.2) 1.1 1 1 44 Physical health problem crisis ?39 (13.4) 315 (12.9) 424 (13.6) 1.1 1 0 45 Criminal legal problem crisis 621 (11.2) 203 (6.3) 416 (13.6) ps0.Family relationship problem crisis 430 (7-6) 212 216 p<0-Job problem crisis 354 (6.4) 191 (7.6) 163 (5.3) ps0.01 (0.5-0.6) 47 Suicide EventiHlstory 4 Left a note 6.466 (34.5) 3.162 (33.6) 3.266 (35.1) 1.1 (1.0?1 .1) 1.2 Disclosed suicide intent 4.405 (23.5) 2.306 (24.5) 2.099 (22.4) p<0-01 0.9 0.9 49 History of ideation 5.990 (31.9) 3.633 (40.6) 2.152 (23.0) ps0.01 0.4 0.4 50 History of attempts 3.?32 (19.9) (29.4) 962 (10.3) ps0.01 0.3 0.3 (0.3-0.3) 51* Alaska. Arizona, Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New 'l?ork. North Carolina. Ohio. Oklahoma. Oregon, Rhode Island. South Carolina. Utah. Vermont. iu'irginia. and Wisconsin. SET Decadent had been identified as having a current diagnosis of mental health problem In coronerimedical examiner or law enforcement reports. 53? Odds ratio re?ects the risk among those without known mental health problem relative to those with known mental health problem. 54 Logistic regression was used to estimate adjusted odds ratio with 95% 1315 after controlling for age, sex. race and ethnicity. Known mental health problem was the 55reference group. includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore. some of percentages forthe diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. 57"? Denominator is decedenis aged 216 years. 53?? Denominator is decedents aged 10?16 years. 59 50 nuscriptcent ra Morbidity and Mortality Weekly Report Page 36 of 36 ?l Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month. 3 Denominator of crisis subgroup is decedents with any crisis within pastor upcoming 2 weeks. Crises depicted here represent the most commonly occurring 4 categories. U1 50 .corn/mmwr Morbidity and Mortality Weekly Report Morbidity and Mortality Weekly Report Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Journal: 1 Morbidity and Mortaiity Weekly Report Manuscript Type: Manuscript CDC author Manuscript ID lcoc-zola-onz I Date Submitted by the Author: 25-Apr-2018 Complete List of Authors: Stone, Deborah; National Center for Injury Prevention and Control, Division of Iviolence Prevention Simon, Thomas; CDC, NCIPC Fowler, Katherine; Centers for Disease Control and Prevention, Division of Violence Prevention Kegler, Scott; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control Yuan, Kerning; National Center for Injury Prevention and Control, Division of Violence Prevention Holland, Kristin; Centers for Disease Control and Prevention, National Center for Injury Prevention and Control Nev-Stephenson, Asha; Centers for Disease Control and Prevention, DVP Crosby, Alexander; National Center for Injury Prevention and Control Jurisdiction - Country [select all that apply): United States Multistate >15 states: I Yes Jurisdiction States and US. Territories (if 15 states, select all that apply}: Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files movies) online. DStone_Figure 1 .emf Page i oni Morbidity and Mortality Weekly Report l-D 4.25.18 Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Sci];1 Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, l?y'lDl Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among people aged 1:10 yea rs, by state and sex, across six consecutive three-year periods (1999-2016], were assessed using data from the National Vital Statistics System for 50 states and Washington, DC. Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 19992016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recentfimpending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with but were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides (15.6f100,000 occurred in the United States (US), among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels (2, Suicide is the 10Th leading cause of death and is one ofjust three leading causes that are increasing (1, 4). Additionally, rates of emergency department visits for nonfatal selhharm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self?harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strategyfor Suicide Prevention{NSSP) (5) calls for a public health approach to suicide prevention with efforts Spanning across multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather, is multi- determined. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal people, providing treatment for mental health problems (MHP) and preventing re-attempts In addition to MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms} among people at risk, and poor coping and problemsolving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, 1 nuscriptcenl ra comfmmun i-D Morbidity and Mortality Weekly Report Page 2 of21 4.25.18 can help reach the nation?s goal of reducing suicide rates 20% by 2025 (7). To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides options for multi-level comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for people aged 210 years only, as determining suicidal intent in younger children can be difficult (3). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Ciassification of Diseases 101". Revision, underlying?cause-of death codes X60-X84, $87.0, 003). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridgedvrace population data releases. National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three- year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronen?medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racei?ethnicity. RESULTS The most recent overall suicide rates {representing 2014-2016) varied four-fold, from 6.9 (D.C.) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +5.53% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (Table 1, Figure 1). Modeled suicide rate trends indicated significant increases for 44 states, for males (34 states) and females (43 states}, as well as for the U.S. overall (Table 1). Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 2? states. While all decedents were predominately male (Table 2; 76.8%) and non?Hispanic white those without known MHP, relative to those with MH P, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5) and racial/ethnic minorities (DR range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI 2.13.8). 2 nuscriptcenl ra cornimmvin Page 3 of 21 Morbidity and Mortality Weekly Report l-D 4.25.18 Among adult decedents, 20.1% of those without known MHP and 15.3% of those with MPH ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.5%] and less likely to die by (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% such as opioids 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP (N29,407j and approximately 85% without known MHP had available circumstances information [Table People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without, versus those with, known MH P, had significantly greater likelihood of any relationship problem/loss (45.1% vs. specifically intimate partner problems (30.2% vs. (12.5% vs. and recently perpetrating interpersonal violence vs. They also were more likely to have experienced any life stressors (54.2% vs such as criminal-legal problems (10.7% vs. or eviction/loss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems were the most common types and did not differ by group. Similarly, among people without versus with MHP, physical health problems (23.2% and 21.4%] and job/financial problems {15.6% and 16.8%) were commonly experienced by both groups. Decedents without known MHP had significantly lower odds of recent release from any institution (a0R=0.5, 95% but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. hospital (43.7% vs. or other facility alcoholfsubstance treatment) 95% than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%] and prior suicide attempts (10.3% vs. More than one in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- female suicide rates between 1999-2014 (3). Additional research into the specific causes of these trends is necessary. Fortunately, data from the 27 states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. 3 nuscriptcenl ra l-D Morbidity and Mortality Weekly Report Page 4 of21 4.25.18 Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention is often oriented towards identification of suicidal people, treatment of MHP and prevention of reattempts. Additional focus on non-mental health factors, further upstream, is essential to a public health approach as the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship problems and other life stressors such as criminal?legal matters, evictionfloss of home, and recent or impending crises. Similarly, people with MHP often experienced relationship problems and other life stressors such as job/financial and/or physical health problems. These findings point to the need to both help people manage the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two?thirds of this group had a history of any mental health andfor substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physicalfmental health problems] Other strategies include creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote helpvseeking, easing transitions into and out of work for people with MHP and other life challenges}, supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, UT) might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders (13), however many methodological variations across studies exist (14). It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known MHP suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices (12] to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. 4 nuscriptcenl ra Page 5 of 21 l-D Morbidity and Mortality Weekly Report 4.25.18 Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No con?icts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 270488-3942 Author Af?liations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Centerfor Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow?Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed], Hindman], Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev-2017-042356. 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo 0. Critical issues in autopsy studies. Suicide Life Threat Behav 2006;36l5l1491-510. Word Count: 1904! 1800 Impsih?mcma nuscriptceril ra comfrrimvvi mummAwM?a kD Morbidity and Mortality Weekly Report Page I5 of 21 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons (Change 1mm Prior Period} Current Overall Overall State Sex State Rate Change 2:32"; 1999 2991 2992 2994 2995 - 2992 2993 - 2919 2911 2913 2914 2915 Flank (State Flank) 1 (State 93mg 1.. Beth 12.311112} 12.2 1+ 9.4} 12- 91+ 9.21 1351+ 9.9} 14.51+ 9.51 1541+ 9.9} 1.5 15 152.911 1122 3.1 91221 25.4 1511115} US. Male 20.911115} 21.21+ 0.4} 21 31+ 0. 01 22.5 2351+ 1.0} 24.5 1+ 1.0} 1.1 113-401) Female 11115} 5.0 1+ 0.3} 3-1-01 1+ 0.4} 6.2 1+ 0.5) 8.9 1+ 2.5 ?15 113-5011 Beth 14.31n2a} 13.41- 9. 9} 14.1 1+ 9. 51 15.51+1.5} 1541+ 9.21 12.51+1.1} 5 15 11:12. 951 25 3.1 131} 21.9 15133} AL Male 25.1 1112a} 23.41.21 24. 4 1+2251+ 1.11 29.1 1+ 1.5} +1.3 151112 951 Female 5.1 11122} 4. 519.31 591+ 9.21 .1.1+1 1} 541+ 9.3} 2.9 1+ 9.2} 2. 5 15 1112. 911 Beth 21.9 1n2e} 24.5 1+ 3.5} 24.25.41- 9.51 23.5 1+ 3.4} 1. 2 15 1112. 951 2 2.51 4} 32.4 15 1131 AK Male 33.2 {We} 35.1 1+ 4.9} 35. 9 1+9. 51 49.1 1+1.2} 49.1 1 9.1} 42.91+2.51 1. 4151 1.112 911 Female 5.5 {n15} 11.4 1+ 2.9} 9.51-1.5} 11.1 1+ 1.2} 9.91- 1.2} 13.2 1+ 3.4} 1.2 15 n25 Beth 12.5 1n2a1 15.5 1+ 9.2} 19.1 1+ 9.5} 19.1 1? 9.91 29. 4 1+ 1. 31 29.9 1+ 9.5} 1.9 15 1112.91) 15 3.1 132} 12.3 15 142} 52 Male 29.3 1n2e} 39.2 1+ 1.9} 39.5 1+ 9.41 39.2 1- 9.51 32. 9 1+ .91 32.4 1+ 9.4} 9.5 15 {1:12.95} Female 2.1 111221 2.5 1+ 9.4} 5.2 1+ 9.21 5.5 1+ 9.5} 21+ 9. 51 9.9 1+ 9.5} 2.2 15 1132.91} Beth 15.5 1n2a1 1551+ 9.3} 1521+ 9.51 12.5 1+ 1.2.9} 2.2 15 1112.911 12 5.2114} 35.5 151151 AB Male 25.21n2a1 25.2 1+ 9.9} 2221+ 9.51 25.21+1.9} 31213.51 33.5 1+ 1.9} 1.5 15 1112.95) Female 5.5 1n2a} 5.9 1+ 9.3} 5.2 1+ 9.41 2.9 1+ 1.2} .5-91.4} 9.5 1+ 2.1} 3.5 15 1112.91} Beth 19.51n1a1 1131+ 9.2} 11.91- 9.3} 12.91+1.9} 11. 519 .11 12.1 1+ 9.3} 9.9 151112.95} 45 1.5145} 14.5 15145} CA Male 12.911151 1541+ 9.5} 12.21? 9.2} 19.1 1+ 1.4} 15. 919 .21 1921+ 9.31 9.5 15 n25 Female 4.1 19151 5.9 1+ 9.9} 4.91- 9.1} 5.4 1+ 95131-91} 5.5 1+ 9.3} 1.2 15 1112.95} Beth 12.31n2e} 19.21+1.9} 19.91- 9.21 29.9 1+ 1.9} 2151+ 1.51 23.2 1+ 1.5} 1.5 15 11:12.91) 5 5.9112} 34.1 151221 CD Male 25.5 1n2a1 39.9 1+ 2.3} 39.51- 9.41 31.5 1+ 1.9} 33. 41+ 1. 91 35.3 1+ 2.9) 1.4 15 1112.911 Female 2.9 91221 5. 2 1+ 1.9.9} 19. 1 1+ 1. 91 19.4 1+ 9.3} 2.5 15 1112.91) Beth 9.591221 951.21 9 1 1+ 9 21 19.21+1.1} 11. 91+ 9.51 11.5 1+ 9.5} 1.5 15 1112.95} 45 1.9143} 19.2 151341 CT Male 15.41n2e} 14.51.51 15 91+ 9.41 15.51+1.5} 1251+ 1.91 12.31- 9.31 9.9 15 n25 Female 3.5 1111a} 3.5 1+ 9.9.2} 9+19.51 5.2 1+ 1.3} 3.5 15 1112.951 Page 7 of 21 Morbidity and Mortality Weekly Report 1 T311011.I 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 4 Age-Adjusted Annual Flate per 100,000 Persons (Change from Prior Period) Current Overall Overall 5 State Sex ?10:13:: E: State Rate Change Eileen-f"; 1999 - 2001 2002 - 2004 2005 2007' 2008 2010 2011 - 2013 2014 - 2016 Flank 5 {State Rank} 1 (State Hgnk) 3 Beth 13.61n1a) 12.21- 1.4) 11 3 1- 0 3) 13.61+ 1.7) 14.2 1+ 0.6) 1441+ 0.2) 0.3 ?In r175 42 0.6 150) 5.3 313150) 9 DE Male 23.0 (ma) 20.3 1- 2.7) 13.3 1-0 .4) 23.1 1+ 3 2) 22.7 1- 0.4) 23.5 1+ 0.6) 0.6 ?11: r115 10 Female 5.3 1n7a) 5.01- 0.2) 601641+ 1.5) 6.21- 0.2) 1.6 34 1175 1; Both 5.31n7a) 6.4 1+ 0.5) 4010.7) 6.3 1+ 0.3) 0.3 34 n15 51 +1.0 146) 16.1 ?79145) 13 DC Male 10.71n7a) 11.1 1+ 0.4) 10.3 1? 0.6) 1271+ 2.4) 10.0 1- 2.6) 11.7 1+ 1.7) 0.3 47.. r175 14 Female 1 7 1n1a) *1 231+ 0.6) T) 3.3 1+ 1.0) 2.6 1? 0.7) 3.6 1+ 1.0) 6?1 0. 6) 3.5 34. r115 15 Both 14.6 We) 15.2 1+ 0.4) 14.3 1? 0.3) 16.3 1+ 1.4) 16.3 1- 0.0) 16. 4 1+ 0.1) 0.6 I34.16305) 23 +1.6 145) 10.6 34146) 16 FL Male 24.31nfa) 24.4 1+ 0.1) 23.6 1? 0.6) 2621+ 2.6) 25.6 1? 0.6) 25. 6 1- 0.1) 0.5 n15 1; Female 6.31n7a) 6.6 1+ 0.5) 6.6 1+ 0.0.5) 61+ 0. 3) 1.4 47.. (112.01) 19 Both 12.31n7a) 1321+ 0.3) 12.31? 0.3) 13. 2 1+ 0. 3) 1371+ 0.5) 15. 0 1+ 1 .3) 0.3 ?in r115 33 2.1 140) 16.2 ?7a 144) 20 GA Male 22.1 (ma) 23.1 1+ 1.0) 21.3 1? 1.6) 21.3 1+ 0. 6) 22.6 1+ 0.7) 2441+ 1.7) 0.5 We 21 Female 5.01n7a) 4.61- 0.2) 4.6 1- 0.2) 51+ 0. 3) 5.6 1+ 0.3) 61+ 0. 6) 2.1 ?34 1p<.05) 2?2 Both 12.3 We) 11.1 1- 1.6) 10.31? 0.7) 14. 51+ 4. 1) 14.41? 0.1) 1521+ 0. 6) 2.0 34;. r175 35 2.4135) 16.3 136) 3: HI Male 20.41n1a) 17.21- 3.1) 15.3 1? 1.3) 21 .31+ 6. 7) 22.5 1+ 0.5) 24. 311+ 1.6) 2.1 34 r115 25 Female 5.4 We) 5.0 1- 0.4) 5.5 1+ 0.5) 7.1 1+ 1 5.) 6.21- 0.1175 26 Both 17.31n7a) 13.2 1+ 2.0) 16.3 1- 0.3) 2161+ 3. 3) 21.3 1+ 0.3) 24.7 1+2 .6) 2.3 33 (p301) 6 7.51 6) 43.2 ?In 1 7) 27 10 Male 26.41n1a) 33.1 1+ 4.7) 31.1 1? 2.0) 34. 3 1+2 3 6) 341132.05) 23 Female 72 (nia) 6.1 1- 1.1) 6.1 1+ 0..1+6 2.3) 4.4 %1p<.05) :3 Both 3.31n7a) 3.61? 0.1) 3.7 1- 0.1) 10. 6 1+2 0 6) 111.0) +1.5 %1p<.05) 44 2.3 136) 22.6 132) 31 IL Male 17.1 1n7a) 16.71- 0.4) 16.21? 0.4) 17.6 1+ 1.4) 16. 511+ 0. 3) 13.6 +1.1 31.113305) 32 Female 3 71n7a) 3.61- 0.0) 3.6 1+ 0.2) 21+ 0. 4) 4. 511+ 0. 4) 21+ 0. 6) 2.4 07.113231) 33 Both 13.0 (nfa) 13.7 1+ 0.7) 1441+ 0.7) 1431+ 0. 5) 16.41+1 4) 17.1 1+ 0. 7) +1.3 %1p<.01) 26 4.1 123) 31.3 34125) 34 IN MaIe 22.4 We) 23.2 1+ 0.6) 24.41+1.2) 2426.3 1+ 1 .6) 1.5 34;. (pc?l) :2 Female 4 61n1a) 5.0 1+ 0.4) 5.3 1+ 0.6.61? 0.2) 2.7 %1p<.01 MD Morbidity and Mortality Weekly Report Page 8 of 21 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:13:59? State Rate Change 1999 - 2991 2992 - 2994 2995 2997 2999 - 2919 2911 - 2913 2914 - 2919 Flank 5 (State Flank) ?7 (State 33,1111 1* Beth 11.511119) 1321+ 141 12.51- 041 14.21+1.41 1551+1?1 +2.1 31 +4.31201 +35.2?191151 IA Maia 20.5 111191 22.1 1+ 1. 51 20.5 1- 1.25.? 1 +1.5 ?19 1p<.051 Female 3 ?111191 1. 01 5.3 1+ 0.51 513.5 ?19 1p<.011 Both 13.31n191 15.1 1+ .51 1551+ 15.31.51 1+ 2. 41 15.41+1.51 2.2 391119.011 15 5.01111 45.0 ?191 51 KS Maia 22.? 111191 25. 01+ 2. 31 2551+ 1.51 25. 51.51 1.5 ?19 1119.011 Female 4.5 1n191.01+ 1.415.? 1 0..51 3.2 ?19 1p<.051 Both 14.1 111191 15.41+1.31 15.? 1+ 1.31 15 21?0 51 15 21+ 2. 01 15.31 1 11 +1.5 20 5.21151 35.5 ?191151 KY Maia 25.0 111191 2551+ 1.51 2551+ 1.41 2?.2 1? 1 .01 30.1 1+ 2. 51 31.?1 1+ 1 .51 1.4 ?191p<.011 Female 4.5 19191 5.2 1+ 0.41 5 1 1+ 0.51 5.1 1+ 0.11 1+ 0511+ 0. 51 3.2 ?19 1119.01) Beth 13.1 111191 12.51? 0.21 134 1+ 0.41 1351+ 0.31 14 41+ 0. 51 1?.0 Tf1+2 .51 +1.5 ?19 19:05) 2? 3.5 1271 25.3 ?191251 LA Maia 225111191 22.31? 0.51 2241+ 0.11 23.3 1+ 0.5) 23 1+ 0. 51 2?.31+ 3.51 1.1 ?19 919 Female 4.5 111191 1? 0.11 5 21+ 0.51 4.5 1? 0.21 5.1 1+ 1.21 1.41 2.5 ?19 19-2551 Both 14.5 111191 13.51? 0.51 14 41+ 0 51 1541+ 1.01 15 51+ 3. 51 15.51? 0.41 2.2 ?19 1119.051 21 4.0 1251 2?.4 ?191251 ME Maia 25.0 111191 22.51? 2.11 2451+ 1.?1 25.? 1+ 1.11 31 1 1+ 5. 41 25.5 1? 1.3) +1.5 ?19 1p<.051 Female 5.31n191 5.31? 0.01 5.21? 0.11 501+ 0.?1 1.51 0.31 3.1 ?19 1p<.051 Both 10.019191 10.3 1+ 0.31 10.1 1? 0.21 10.21+ 0.11 10 1+ 0. 51 1051+ 0.11 0.5 ?191p<.051 4? 0.5 145 5.5 ?19 145 ?511 MD Maia 1?.5 111191 1?.51+ 0.11 1?.31? 0.51 0.41 15. 2 1+ 0. 51 15.01? 0.21 0.2 ?19 1115 Female 3.5 111191 3.5 1+ 0.4) 3.5 1+ 0.01 1? 0.21 4.1 1+ 0.41 4.5 1+ 0.4) 1.3 ?19 1p<.051 Both 4 19191 1+ 0.1.01 5.5 1+ 0.4) 1001+ 0.31 2.3 ?19 1119.01) 45 2.5 134 ?1 35.3 ?19 120 ?1 MA Male 12.1 (1119) 1251+ 13 31+ 0 51 1541+ 2.11 15.21? 0.21 1501+ 0.51 2.0 ?191p<.011 Female 3.3111191 2.51? 0.41 4 01+ 1.01 3.51? 0.11 451+ 1.01 4.51? 0.21 3.0 ?19 1.99.051 Both 11.5 111191 1251+ 0.?1 1251+ 0.41 1351+ 1.01 1451+ 0.?1 1551+ 1.11 +1.5 ?19 1p<.011 33 3.5 1251 32.5 ?191241 MI Male 20.0 111191 2051+ 0.51 2151+ 0.?1 2251+ 1.31 2351+ 1.01 2501+ 1.21 1.5 ?19 1119.01) Female 4.4 19191 4.5 1+ 0.41 5 0 1+ 0.21 5.5 1+ 0.51 5.5 1+ 0.31 1+ 0.51 2.5 ?19 1912.011 Page 9 of21 Morbidity and Mortality Weekly Report 1 3 Table 1. Trends in Suicide Rates among Persons 2 10 Years of l?lge,r by State and Sex, National Vital Statistics System, 1999 2016 4 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall 5 State Sex ?10:13:59? State Rate Change 3 1330 - 2001 2002 - 2004 2005 - 2007 2000 - 2010 2011 - 2013 2014 - 2010 Hank 5 (State Flank) ?7 (State 3* 3 Both 10.? (1113) 11.5100) 1241+ 0.8) 1201+ 0.5] 14.21+1.3) 1501+ 0.91 2.3 ?11: (1:14.01) 38 4.31131 40.0 ?/31 81 9 MN Maia 13.3 1117a) 13 3 1+ 1.11 2041+ 1.01 2031+ 0.31 2231+ 1.31 2331+ 0.41 +1.7 31, 1132.011 10 Female 3.3 1n7a1 21 0. 31 4 31+ 0.31 5.1 1+ 0.4) 5.3 1+ 0.31 331+ 1.21 4.2 31, 1114-011 1; Both 12.311131 14 1 1+ 21 1471+ 0.31 1551+ 0.31 1531+ 0.11 1521- 0.31 1.1 331112.051 33 2.3 1331 17.3 31, 1401 13 M3 Male 22.3 1n7a1 24. 3 1+ 1. 71 25.1 1+ 0.31 23.3 1+ 1.71 25.31- 0.31 25.3 1- 0.31 0.7 31. n73 14 Female 4.3 1n7a1 .010 .71 551+ 0.51 5.5 1- 0.01 3.4 1+ 0.31 3.2 1- 0.21 2.4 3.4 1114.011 15 Both 14.71n7a1 14.1 1-0 31 1541+ 1 31 1301+ 0.71 1731+ 1.71 2001+ 2.31 2.2 31+ 1114.011 13 5.3 1151 33.4 341171 ?5 Mo Maia 25.31n7a1 23.71.31 2531+ 1 31 2331+ 1.01 2331+ 2.31 3221+ 3.31 1.3 331112.051 1; Femaie 5.41n7a14+1 0.11 31 1+ 0.71 3.3 1+ 0.21 741+ 1.11 331+ 1.21 3.2 0.1. 1112.011 19 Both 21.1 111751 22. 3 1+ .41 2331+ 1 01 2471+ 1. 11 2371+ 2.01 2321+ 2.51 2.1 31. 1134.011 1 3.0 1 21 33.0 331111 20 MT Maia 33.3 1117a} 7 3 1+ 0. 41 3331+ 2.51 33 7 1-0 41.0 1+ 1.41 4551+ 4.41 1.3 31. 1112.01) 21 Female 3 7 (Ma) 41+ 1. 31 3.41- 0.11 10 01+ 1 .31 1231+ 2.31 13.1 1+ 0.51 4.3 31. 1114.011 22 Both 1271117131 1221- 0.51 1231+ 0.41 11 .7-10 .31 1351+ 1.31 14.31+1.31 +1.0 31. n75 40 2.1 1421 +132 37.: 1431 3: NE Maia 22.2 111751 20.71- 1.51 20.3 1- 0.41 13. 3 1-0 .51 22.0 1+ 2.21 2331+ 1.31 0.3 31. 111's 25 Female 3.3 1n7a1 4.2 1+ 0.41 5.1 1+ 0.31 0-11.11 5.5 1+ 1.41 531+ 0.31 2.3 31+ n75 23 Both 23.3 1117a) 2231- 0.31 22.1 1 0.51 22.13 0. 51 2141-121 23.1 1+ 1.0.21511 - 1.0 341511 27 NV Male 38.3 (1113) 36.? 1? 1 35.1 1? 1.6) 35. 0. 5) 32.5 3.01 3541+ 2.3) ?7?5 n13 23 Female 3.31n7a1 351+ 0.51 331+ 0.11 10 01+ 0. 41 1031+ 0.31 1121+ 0.31 1.5 31, 1114.011 :3 Both 13.5 1117a} 1251 1.01 1331+ 0.31 15. 2 1+ 1 .31 1531+ 0.31 2001+ 4.21 2.7 33 1114.051 17 3.5 1 31 43.3 341 31 31 NH Maia 22.51n7a1 21.1 1- 1.41 2171+ 0.31 24.31.11 2541+ 0.31 3031+ 5.21 2.2 331112.051 32 Female 5.3 1n7a1 4.31? 051 531+ 1.012+1 0. 41 3.3 1+ 0.41 331+ 3.21 3.3 31, 1112.051 33 Both 7.31n7a1 7.71- 0.11 7.51- 0.21 .01+ 0. 51 331+ 0.31 321+ 0.41 +1.3 31. 1152.051 50 1.5 1471 13.2 3.411351 34 NJ Male 13.01n7a1 13.1 1+ 0.01 12.31- 0.51 13.71+ 1.11 1451+ 0.31 1431+ 0.11 0.3 31+ 1114.051 :2 Female 3.2 1n7a1 2.31? 0 31 301+ 0.01 3 1- 0.11 3.3 1+ 0.31 4 41+ 0. MD Morbidity and Mortality Weekly Report Page 100f21 Table 1. Trends in Suicide Rates among Persons 2 10 Years of A96, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons 1Change from Prior Period) Current Overall Overall State Sex ?10:13:: :3 State Rate Change 1999 - 2991 2992 - 2994 2995 2997 2999 - 2919 2911 - 2913 2914 - 2919 Flank 5 (State Flank) (91911.- an111 1* Both 22.011119} 22.01? 0.1} 21.31? 0.2} 23.01+ 1.2} 24.1 1+ 1.1} 26.01+ 1.3} 1.1 319 1p<.05} 4 4.0 124} 13.3 31.133} NM Male 36.3 11119} 37.7 1+ 0.3} 36.4 1- 1.2} 35. 3 1?0 37.1 1+ 1.3} 40.71+3 0.4 ?11. 1119 Female 3.511119} 7.41? 1.1} 3. 2 1+ 0.7} 10. 71+ 2. 6} 1171+ 0.3) 12. 0 1+ 0. 3} 3.3 1p<.05} Both 7.211119} 7.1 1? 0.1} 71+ 0.6} .4+10 351+ 1.1} .3?01 2.1 391119.01} 43 2.1 141} 23.3 34.127} NY Male 12.511119} 12.21? 0.3} 12. 3 1+ 0.7} 13 31+ 1.0} 1541+ 1.4} 14. 510 3} 1.4 319 1p<.05} Female 2.711119} 2.61? 0.1} 01+ 0.3} 51+ 0. 5} 421+ 0.7} 61+ 0 5} 4.2 ?19 1132.01} Both 13.6 (1119} 13.51? 0.1} 1371+ 0.1} 14 21+ 0. 5} 1451+ 0.4} 15. 3 1+ 0. 3} 0.3 1119.01} 34 +1.7 (44} 12.7 33147} NC Male 22.711119} 22.7 1+ 0.0} 22.21? 0.6} 23. 3 1+ 1. 1} 2331+ 0.0} 23.31+ 0. 6} 0.4 ?19 1119 Female 5.6 1n190.3} 6.7 1+ 0.7} 61+ 0. 3} 2.0 34.. 1119.05} Both 13.311119} 14.61+1.3} 1601+ 1.4} 16 61+ 0. 6} 13.41+1.3} 20. 31+2 5} 2.3 33111901} 14 7.6 1 5} 57.6 ?/91 1} ND Male 21.4 111.19} 24.6 1+ 3.2} 2301+ 3.4} 27 1 1?0 23.6 1+ 2.5} 32 71+ 3. 0} 2.5 ?11. Female 5.6 11119} 4.51? 1.0} 3.7 1? 0.3} .7+1 2. 0} 6.7 1+ 1.0} 51+ 1 3.3 1115 Both 11.611119} 1231+ 0.3} 13.1 1+ 0.3} 13 41+ 0. 2} 14.31+1.4} 1531+ 1.0} 2.0 31.. 1119.01} 32 4.2121} 36.0 113} OH Male 20.411119} 2031+ 0.5} 2221+ 1.3} 22 1 1?0 2421+ 2.1} 25. 51+ 1 1.5 31. 19:01} Female 4.0 1n19} 4.7 1+ 0.7} 31+ 0.1} 3+1 0. 5} 6.2 1+ 0.3} 71+ 0. 6} 3.4 ?11. 1p<.01} Both 17.011119} 16.51? 0.6} 17.21+ 0.3} 13. 4 1+ 1.1} 2071+ 2.3} 23 512.3} 2. 3 ??19 11:19 05} 7 6.4 110} 37.6 39112} UK Male 23.5 11119} 27.3 1? 1.2} 2731+ 0.5} 30 3 33.4 1+ 3.?13 1139. 05} Female 6.611119} 6.41? 02} 751+ 1.1} 10?015} 3.51+ 1.6} 10 31+ 1 2. 3 ?1o 1p<. 05} Both 16.4 (1119} 17 7113} 17.71? 0.0} 13 16+ 0. 3} 13.31+1.2} 21 1 1+ 1 +1.6 31.1119 01} 13 4.6113} 23.2 319123} Male 27.411119} 23512.1} 23.51? 0.3} 23 .5 1+1. 0} 31.41+1.3} 33.01+ 1.6} +11 11:19. 01} Female 6.5 11119} 7.1 1+ 0.6} 7.7 1+ 0.6} .4+01.7} 3.3 1+ 0.4} 31+ 0. 3} 2. 7 391 1.119 01} Both 12.1 11119} 1251+ 0.4} 1231+ 0.3} 13 31+ 11} 1501+ 1.1} 16. 3 1+ 1.2} 2.0 ?13 1112.01} 30 4.1 122} 34.3 a111121} Male 21.0 (1119} 21. 31+ 0.3} 2131+ 0.6} 231 2471+ 1.7} 26.1 1+ 1 +1.5 33111901} Female 4.211119} 4 6 1+ 0. 3} 461+ 0.0} 4+01.3} 601+ 0.6}11+ 1.1} 3.5 391119.01} Pagei'l of21 MD Morbidity and Mortality Weekly Report Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:35? State Rate Change 1000 - 2001 2002 - 2004 2005 - 200? 2000 - 2010 2011 - 2013 2014 - 2010 Flank 5 (State Hank) ?7 (State Both 9.411110) 9.01? 0.3) 9.010 0) 12.313 3) 11.91? 0.9) 12 3 1+ 0. 2.3 31: 1p<.05) 43 3.2130 34.1 33123 RI Maia 15.41n1a) 1521? 02) 1431?0 .3) 21 21+ 3. 4) 19.21? 2.0) 19 3 1+ 0. 4) 2.2 34.. n15 Female 4.0 (Na) 3.3 1? 3.3 1+ 0.4) .11+ 1 .3) 5.1 1+ 0.0) .11+ 1 .0) ?11. 1p<.05) Both 12.31012) 1301+ 0.2) 13.? 1+ 14. 91+1.2) 13.01+1.1) 1+ 1 2.3 ?14. {pi-<01) 23 4.9 11?) 33.3 34.110) SC Male 21.3 111.12) 2251+ 1.2) 22.31? 0.1) 24. 312.2) 23.1 1+ 1.5) 3 01+ 1.9) 1.3 1p<.01) Female 5.4 1n1a) 1? 301+ 1.3) 21+ 0. 2) 1+ 0.3) .41+ 1 4.) 3.4 113-505) Both 15.?1nia) 1531+ 0.1) 1?.1 1+ 1 19.31 2.2) 19.? 1+ 0.4) 2231.9) 2.5 3911:1301) 10 ?.01 44.5 34: 1 3) 30 Male 27.311110) 2331?13) 2?.9 41+ 1. 3) 30.1 1+ 2.2) 3201+ 1.9) 33. 3 1+ 1 .3) 1.3 1p<.01) Female 4.211110) 531+ 1.3) 1+ 0. 3) 3.3 1+ 2.0) 1? 1.0) 1131.0) 5.3 31; 1p<.01) Both 14.311112) 1521+ 0.3) 13.141+ 0.3) 1?.21+ 1.1) 1?.21+ 0.0) 13. 2 1+ 1.0) +1.4 ?19 1p<.D1) 22 3.5 123) 24.2 ?3131) TN Mate 25.1 (Ma) 25.4 1+ 0.3) 2331+ 1 3) 23. 0 1+ 1.2) 23.3 1+ 0.3) 29. 3 1+ 1 .2) 1.2 ?11. 1p<.01) Female 5 4 11112) 3.3 1+ 0.9) 1+ 0.4) 5+1 0.3) 3.91? 0.3) 31+ 0. 1.9 10-505) Both 12.2 111.12) 12.? 1+ 0.3) 12.31? 0.4) 13.21 0. 9) 1331+ 0.3) 14. 1+5 0. 9) 1.1 91.. 1.02.01) 41 2.3 13?) 13.9 133) TX Maia 20.4 11112) 20.9 1+ 0.5) 20.4 1- 0.3) 22.0 1+ 1 .3) 22.2 1+ 0.3) 23.1 1+ 0. 9) 0.9 1p<.05) Female 4 3 (n12) 5.4 1+ 0.3) 5.0 1? 0.4) 201+ 2) 5.3 1+ 0.4) 41+ 0. 3) 1.3 ?11. Both 1? 2 (r112) 1901+ 1.3) 1321? 20. 21+ 2. 0) 2401+ 3.3) 2521+ 1. 2) 31010301) 5 3.0 1 3 W) 43.5 31.1 4 W) UT Male 23.21012) 31.1 1+ 2.9) 29.41?1.?) 32.1 1+2 3?.31+ 33. 0 1+ 0. 2) 2.1 ?19 1132.05) Female 3.31012) 0.3) 0.1) 1031+ 2.1) 1231.0) 4.4 ?20 113-501) Both 13.2 We) 1321+ 3.0) 14.9 1? 1.3) 13 13.? 1+ 2.1) 19. 1+ 1 0) 2.4 31,102.01) 13 3.41 9) 43.3 34.1 2) VT Male 23.31n1a) 23. 3 1+ 4. 3) 24.31? 4.0) 3131.01+ 32 51+ 1.5) 1.9 102.05) Female 4.3 11112) 5.2 1+ 0.9) 3.4 1+ 1.3) .313.3 1112.01) Both 12.311112) 12. ?10.1) 1291+ 0.3) 13. 31+ 0. 1431+ 0.1132.01) 3? 2.2 139) 1?.4 34.141) VA Male 21.3 (Ma) 21. 31? 0.2) 21.01? 0.4) 22. 51+ 1. 5) 2331+ 1.2) 23.9102) 0.9 31. 1p<.05) Female 5.3 (nfa) 521? 0.1) 591+ 301? . 3) 3.4 1+ 0.3) 91+ 0. 5) 1.3 1p<.05) 005401015de N-D Morbidity and Mortality Weekly Report Page 12of21 Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Flate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex ?10:13:: :1 State Rate Change 1999 - 2001 2002 - 2004 2005 - 2007 2008 - 2010 2011 - 2013 2014 - 2016 Flank 5 (State Flank) ll (State Rink) 1+ Both 14-81n1a) 15.410 5) 14.81? 0.6) 15.?1+ 0. 9) 16 61+ 0.9) 1161+ 1.0) +1.1 ?11; 1p<.05) 24 2.8133) 18.8 51:13?) WA Male 24.?1n1a) 25. 21+0 .5) 24.1 1? 1.1) 25.1 1+ 1.0) 26 01+ 0. 9) 2?.1 1+ 1.1) 0.6 Female 5.9 (nta) 41+ 0. 6) 6.2 1? 0.2) 9+10 71+ 0. 8) 851+ 0.8) 2.5 ?to (15:91) Both 15.61n1a) 1121+ .6) 1631-05) 16?001.?) 19.21+3.2) 21.41+ 2.2) 1.8 n/s 11 +5.8113) +371 819114) WV Male 27.2 (Ma) 30 1 1+ 2.9) 2851?15) 2161.0) 51+ 3.9) 3351+ 2.0) 1.1 ?to nis Female 5.3 (nra) 51+ .1) 81+ 0.2.2) Both 13.1 (nta) 1351+ 0.4) 14.0.1+05) 15.101+ 1.0) 1531+0.3) 1651+ 1 2) 1.5 ?to 1p<.01) 28 +3.4129) +258 [343130) WI Male 21.?1n1a) 22. 21+ 0.5) 22. 7? 1+ 0. 5) 24. 01+ 1.2) 2441+ 0. 4) 25.? 1+ 1 3) 1.1 Gale {pi-:01) Female 5.1 (nta51+ 0.1) 7151+ 1.0) 2.5 1p<.01) Both 20.?1n1a) 23.14 .7) 22. 51?0 9) 2541.8) 28. 9 1+ 3. 5) 28.81? 0.1) 2.3 ?11. 1p<.01) 3 8.1 1 1) 39.0 ?/21 9) W?r? Male 34.8 (Ma4151.2) 4?.1 1+ 5. 6) 44.61? 2.4) 1.8 ?to 1p<.05) Female (nla) 8.2 1+ 0.6) 9.2 1+ 0.9) .41+0 .2) 10. 1+ 1. 4) 12.6 1+ 1.9) 3.2 fro 1p<.01) Ftates are age-adjusted to the year 2000 standard. Model-estimated average annual percentage change based on all reporting periods: p-value indicates statistical significance of trend: n1s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 - 2016. Hanks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically significant difference. 'l Overall rate change is between the first (1999 - 2001) and last {2014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first 11999 2001) and last {2014 2016) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). ll Ftate based on 20 suicides. Differences between ranks do not necessarily imply a statistically significant difference. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Page 13 of2?r mumm-h-wM?s ND Morbidity and Mortality Weekly Report Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted {95% (n=20,446} prob em* Mental Health Square (95% c? a} {$9,407} Problem {n=11,039) Sex Male 15302063) 6,469l683} 9,233l83.6) p<.01 Female 4,744l232) 2338812} 1,306l16.4) p<.01 Age" 10?24 2,804l13.7) 1,211l12.9} p<.01 25?44 6,456l31.6) 3,036l323} p<.05 45-64 7271887.?) 3,820i40.6] 339885.31 p<.01 65+ 3,468l17.0] 1,340l14.2] 2,128l19.3] p<.01 Race/ethnicity White, non?Hispanic 17,102l83?) 8,165l86.8} 8,937l810) p<.01 Black, non-Hispanic 1,228l5.0] Mil-4.4) 81347.4} p<.01 American Indiaanlaska Native, non-Hispanic 378l1.8} 112(1.2) 2669.4) p<.01 Asian, non?Hispanic 576l2.8} 235(2.5) 3418.1) p<.05 Hispanic 463(4.9) 6335.7} p<.05 Other 66(03) 21(02) 4503.4) p<.05 Extended demographics Ever served in military? 3,429l17.8) p<.01 Homeless 240l1.2} 104(11) 136{1.3) Incident Type Single suicide 20,063l982) 9318(991] p<.01 Homicide followed by suicide 319l1.6} 64(07) 255{2.3} p<.01 Multiple suicides 64(03) 25(03) 39(0.4) Method Firearm 93091435) p<.01 5,907l28.9] 2,940i313} p<.01 Poisoning 1,861l19.8] p<.01 mumm-h-wM?u ?0 Substance class causing death?? Other over-the-cou nter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for 2 1 substance? Substance detected Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive Morbidity and Mortality Weekly Report 94431.4} 524(203} 219(73} 931304.41 4,442l40.6] 8,554l41.8) 499(53} 75151372) 73597} 55591311) 2,214t4os) 555(3ss) 508(32.7) 544(34s) 458(251) 195(105) 730(33) 5553005} 5,409i57.5} 4,253r453} 1,238i29.1} 1,6398%} 3,866i41.l} 215(55) 375(102) 719(22.7) 3,103l33.0} 355(31.1} 336(29.4} 155(137} 155(13.7) 24(2.1) 315{7.4} 4295355) 325(212} 283(53) 3319855] 360{9.2} 75103.1} 479(205) p<.01 p<.Dl p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 ones-0.7) 0.703.703) earns?1.1) cares-0.4) 0.30.1433) 0.703.507) Dalila?1.0) news?1.0) 0.503.507) Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan, Minnesota. New Hampshire. New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma. Oregon, Rhodc Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Page 14of21 Page 15 of 21 Morbidity and Mortality Weekly Report Decedent had been identi?ed as having a current diagnosis of mental health problem in coronen?medieal examiner or law enforcement reports. i5 Odds ratio refleets the risk among those without known mental health problem relative to those with knewn MHP. 1' Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age. sex, race and ethnicity. Known MHP was used as the reference group. Decedents were aged [0 years and older, as per standard in the suicide prevention literaturedecedents aged 13 years of age and older reported military service status. Denominator is decedents who died by poisoning. including overdose. Denominator is with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. 43-h- new mumm-lh-wM?n ND Morbidity and Mortality Weekly Report Page 160f2i Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted on? Problem+ Mental Square [95% [95% cu Heakh Problem Suicide with known circumstances 18.764813) 9.407(100] p<.Dl Mental Health Am; Current Diagnosed Mental Health Problem? 7.076(752) Anxiety disorder Bipolar disorder Schizophrenia 50915.4} PTSD ADDKADHD 22642.4} Unknown 760l3.1} Current depressed mood 3,962l42.1) 3,076l32.9) p50% observed in 25 states (Supplementary Table; 53?8 5) (Figure). Modeled suicide rate trends indicated significant increases in 44 states, among males (34 states) and Females (4.3 states), as well as For the United States overall (Supplementary Table: Nationally, the model? estimated average annual percentage change For the overall Alaska. Arizona. Colorado, Connecticut. Georgia, Hawaii, Kansas, Kentucky, Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. Newjetsey, New Mexico. 1hit-.w York. North Camlina, Ohio, Oklahoma, Oregon, Rhode Island. South Carolina, Utah, Vermont, Virginia. and Wisconsin. US Department of Health and Human Servicesr?Centers for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 I I Increase: increase: 19%?30% Increase: 5%?18% Decrease: 1.0% Per 100,000 population, age-adjusted to the 2000 U5. standard population. suicide rate was an increase By sex, estimated national rate trends Further indicated significant average annual per? centage change increases For males and Females (Supplementary Table; Suicide decedents without known mental health conditions (1 1,039: 54.0%} were compared with those with known mental health conditions 40.0%) For 27" states. Whereas dece? dents were predominantly male (Table 1) and non: Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 2.3, 95% CI 2.2+25} and belong to a taciali?ethnic minority (OR range 1.2-2.0l. Suicide decedents without known mental health conditions also had significantly higher odds onerpetrating homicide Followed by suicide 2.9, 95% CI Among decedents aged 218 years, 20.1% oFthose without known mental health conditions and 15.3% oF those with mental health conditions had previously served in US. military or were serving at the time oF death. Whereas firearms were the most common method Suicide overall decedents without known mental health conditions were more likely to die by Firearm and less likely to die by (26.99/43) or poisoning (l than were those with known mental health conditions 31.3%, and 19.8%, respectively). These difFerences remained signi?cant in the adjusted models. 'l'oxicology testing was less likely to be perFormed For dece- dents without known mental health conditions. Among those with toxicology results, decedents without known mental health MMWR June 8,2008 Vol.6? No.22 E119 TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental Morbidity and Mortality Weekly Report health conditions National Violent Death Reporting System, 21' states,* 2015 Known mental No known mental Total health condition1 health condition 20,446] in 9.407] [n 1 1,1139} Chi-square 0R5 Adjusted on'" Characteristic 00.1%] [110.1%] 010.1%] p-vaiue {95% Cl] [95% Sex Male 15.202 {76.81 6.459 (58.81 9,233 (83.61 {0.01 2.3 NA Female 4,244 {23.2] 2.933 (31.2] 1,306 (16.4] <0.01 0.4 NA Age group [yrsW? 1044 2,304 1,211 (12.9] 1,593 (14.4] <0.01 1.1 (1.1-v1.2] NA. 25?44 5.4563115] 3,036 (32.3] 3,420 (31.0] <0.05 03103?10] NA 45?64 {31.11 3.320 (?10.61 3.898 (35.3] {0.01 0.13 NA 265 3,463 {17.0} 1,340 (14.2] 2,128 (19.3] (0.01 NA RacefEthnicity White, non?Hispanic 12,102 (83.6] 8,165 (86.8] 8,93? (81.0] (0.01 0.5 NA Black, non-Hispanic 1,225 411 81213.4] <0.01 1.7 NA American Indianfnlaska Native, 37811.3] 112 26612.41 40.01 2011.54.61 NA non-Hispanic Asian, non-Hispanic 525 235 341 {0.05 1.2(1.141.5] NA Hispanic 1,096 463 533 <0.05 NA Other 00 21 45 (0.05 1.3 NA Extended demographics Ever served in militantll 3,429 {113] 1,354 (15.31 2.025 (20.1] <0.01 1.4 1.1 Homeless 24011.21 10411.1] 136 {1.31 NS 1.1 1.2 {0.51?1.51 Incident type Single suicide 20,063 (98.2] 9,313 (99.1] 10,?45 {9714] {0.01 0.3 0.4 Homicide followed by suicide 319 6410.7] 25512.31 error 3.5 2392-381 Multiple suicides 64 25 3901.4] NS 1.3 (0.32.21 1.6 Method Firearm 9,909 {43.51 3,321 {40.5} 6,088 {55.31 (0.01 1.3 {tr-1.9} 1.511.543) 5,907 (23.9] 2,940 {31.31 2,96? [26.9] <0.01 013018?09] 0.8 [0.1503] Poisoning 3,003 (14.2] 1,351 (19.3] 1,142 (10.4] {0.01 0.5 0.5 Substance class causing Other ovar?the?counter] 1,021 {34.0] 05:61:35.8] 355 (31.1] {0.01 0.8 0.9 Upioids 944 {31.4] 608 330 (29.4] NS 0.9 0.9 Antidepressants 300 {26.61 1544 {34.61 1515 {1 3.?1 (0.01 0.3 0.3 Benzodiazepines 624 (20.3] 458 (25.1] 156 [13.1] 4:001 0.5 0.5 219 19511105] 24 <0.01 0.2 0.2 (ill?0.3] Other 1.595013] 130 (8.31 815 {0.05 0.9(034101 0.21 {0.13?1.01 See table footnotes on next page. conditions Were less likely to tesr positive for any substance overall 0.8, 950/0 Ci including opioids 0.90, 95% CI 0.814199). but were more likely to test positive For alcohol 1.2, 95%, CI information on circumstances surrounding suicide were available For all decedents with mental health conditions (9,407) and approximately 35% oF those without known mental health conditions (9,357) in 27 states (Table Persons without known mental health conditions were less likely to have any problematic substance use 0.7, 95% CI 0.7?0.8] than Were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or subStance use treatment just over half were in treatment at the time of (leads. Decedents without known mental health conditions had a signi?cantly higher likelihood ofany relationship problem/loss 520 MMWR June8,2018 Vol.6? No.22 than did those with known mental health conditions (39.00.10), specifically intimate partner problems (30.2% ver? sus {17.5% versus and perpetrating interpersonal violence in the past month versus Decedents without known mental health condir tions were also more likely than were those with known mental health conditions to have experienced any life stressors (50.5% versus 412%} such as recent criminal legal problems (10.7% versus or evictionr'loss ofhome versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks. respectively] crisis (a current or acute event thought to contribute to the suicide) (32.9% ver- sus 20.0%] . All of these differences remained significant in the adjusted models. Physical health problems and jobffinancial problems were commonly contributing stressors among both persons without mental health conditions (23.2% and 15.0%, respectively} and those with mental health conditions (21.4% US Department of Health and Human Servicestente-rs for Disease Control and Prevention Morbidity and Mortality Weekly Report TABLE 1. {Continued} Selected demographic and descriptive characteristics ofsuicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health condition'r health condition 20,446] {n 9,413?) {n 11,039} Chi-square 0115 Adjusted oFt?1 Characteristic 110.1%} No. 11%} No. i?itrl p-value {95% Cl} {95% Cl} Toxicology results Any toxicology testing 1 3,317 {65.1} 6,658 {20.3} 5,659 {50.3} (0.01 0.6 0.7 {0.15-0.20 Positive for 21 substance? 9,913 {24.4} 5,192 {18.0} 4,221 {20.9} ?0.01 0.3 Substance detected*? Alcohol Tested 10,950 {53.6} 5,409 {57.5} 5,541 {50.2} (0.01 0.8 Positive 4,442 {40.6} 2,115 {39.1} 2.32? {42.0} c001 1.1 1.2 Upioids Tested 8,554 {41.5} 4,253 {45.3} 4,296 {38.9} ?10.01 0.8 0.8 Positiva 2,279 {20.6} 1,2 38 {29.1} 1,041 {24.2} <0.01 0.8 0.9 Benzodiazepines Tested 3.124 {39.2} 4,226 {44.9} 3,398 {35.3} {0.01 0.7 Positive 2,464 {30.3} 1,639 {38.3} 325 {21.2} c0111 0.4 0.5 Cocaine Tested 2,973 {39.0} 3,866 {41.1} 4,112 {312} <0.01 0.9 0.9 Positive 499 210 233 (0.05 1.2 {1 1.2 AmphetaminES Tested 11,61 5 {312} 3,696 {39.3} 3,919 {35.5} ?:001 0.9 0.9 Positive 730 326 {10.2} 300 NS 0.9 1.0 {0.8?1 Marijuana Tested 6,569 {32.1} 3,127 {33.2} 3,442 {31.2} (0.01 0.9 0.9 Positive 1,471 {22.4} 3'10 {22.70 2'61 {22.1} NS 1.0 0.9 Antidepressants Tested 5,425 {25.5} 3,103 {33.0} 2.322 {21.0} {0.01 05 0.6 Positive 2,21 4 {40.3} 1,735 {55.9} 429 {20.6} (0.01 0.2 0.2 Abbreviations: CI con?dence interval; NA not adjusted: N5 not signi?cant; OH odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah. Vermont, Vi rginia, and Wisconsin. 1 Decedent had been identi?ed as having a current diagnosis of mental health condition in coronerrmedical examiner or law enforcement reports. 5 DR reflects the risk among those without known mental health condition relative to those with known mental health condition. 1? Logistic regression was used to estimate adjusted 01-1 with 95% after controlling for sex, age group. and racefethnicity. Known mental health condition was used as the reference group. ?i Decedents We're aged 210 years, as per standard in the suicide prevention literature. ii Denominator is decedents aged 213 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. 11'? Denominatoris decedents with any toxicology testing. Denominator for each positive group is the number tested for the substanca in that group. and 16.8%, respectively}. Similarly, among all persons with recent crises. intimate partner problems were the most com- mon types and did not by group. Decedents withour known mental health conditions had signi?cantly lower odds of recent release From any institu? tion laOR 0.5, 95% C1 Among those recently released, decedents without known mental health conditions were significantly more likely than decedents with mental health conditions to have been released from a correctional Facility (25.7% versus hospital {433% versus or other facility, such as an alcohollsubstance use treatment facility {24.2% versus Among decedents with known mental health conditions who were recently released from an institution1 46.7% were released from facilities. Decedents without known mental health conditions were significantly less likely to have a history of suicidal ideation US Department of Health and Human Services/Centers for Disease Control and Prevention {25.00.41} or prior suicide attempts compared with those with known mental health conditions {40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% ofpersons without and with known mental health conditions, respectively. Conclusions and Comments During 1999?2016. suicide rates increased signi?cantly in 44 states, and 25 states experienced increases :309/0. Rates increased signi?cantly among males in .34 states, and females in 43 states. Additional research into the specific causes of. these trends is needed. Data from the 27 states participating in provide important insight into cir- CLIrnstances surrounding suicide and can help States identify prevention priorities. MMWR I June 8,2018 Vol.6? r? No.22 i321 TABLE 2. Circumstances preceding suicide among decedents aged 21]] years with and without known mental health conditions National Morbidity and Weekly Report Violent Death Reporting System, 27 states,* 2015 Known No known mental health mental health Chi? Total conditior?' condition square 005 Adjusted Characteristic No. No. 110.1%] p-value (95% Cl] (95% Cl] Suicide with known circumstances 13.7641913] 9,407 (100] 9.35? (84.0] ?10.01 NM. NM Mental health Any current diagnosed mental health condition? ??11 1076 0?52] NM WA NM WA Anxiety disorder ?11 1,529 (16.3] NM. NM (WA NIP. Bipolar disorder ?11 1,431 (15.2] NM NM MIA NM Schizophrenia ~11 509 NM NM NIP. NM PTSD 424 MM we no. NM _11 225 {2.41 mm NM NM NM Not speci?ed 16010.1] N14. WA WA NRA Current depressed mood?? 2,033 (37.5] 3,962 (42.1] 102682.91 (0.01 0.1" 0.2 Problematic substance use Any 5,319 (23.3] 2,976 (31.5] 2,343 (25.0] {0.01 0.7 (or-0.3] Alcohol 3.268 (12.4] 1,862 (19.0] 1,406 (15.0] (0.01 0.2 0.2 Other 3,084 (16.4] 1,268 (113.8] 1,316 (14.1] {0.01 0.2 0.7 Treatment Current mental healthr?substance use treatment 5,141 5,02? (54.0] 64 <00] 0.01 (0.01?0.01] 0.01 (0.01?0.01] Ever treated for mental disorder 6,712 (35.3] 6,323 (67.2] 394 <00] 0.02 (0.02?0.02) 0.02 (0.02?0.03] Relationship problemsfloss Any relationship problemfloss 2.9430114] 12261396] 4,222 (45.1] {0.01 Intimate partner problem 5,098 (22.2] 22201241] 232880.21 <0.01 1.4 1.4 (1.3?1 Perpetrator ofinterpersonal violence in past month 414 131 283 :00] 2.2 2.0 lir'ictim of interpersonal violence in past month 84 53 31 <0.05 0.6 (04?09] 0.8 (0.5-4 Family relationship problem 1,671 323 293 NS 03013?10} 1.0 (0.9?1 Either relationship problem (nonintimatei 403 202 201 NS 1 0 (0.3-1.21 1.1 Argument or conilictlnot speci?ed] 2.914(155] 1.2281156] 1,636 <00] 13(12?15] 1.4 Death of a loved one {any} 1,497 326 {0.01 0.3 0.9 (08?09] Nonsuicide death 1,181 547 534 (53] <00] 0 8 09013?10] Suicide of family or friend 31912.0] 211'123] 162 (0.01 1' (06-03} 0.8 Other life stressors Anylife stressor 9,1?1 (48.9] 44420112] 4,129 (50.5] <00] 1.1 1.1 Recent criminal legal problem 1,530 586 1.002 <00] 1.0 (Lo-2.0] 1.7 (15-10] Other legal problem 248 328 320 NS 1.0 (0.8?1 1.0 Physical health problem 2,012 (21.4] 2,16? (23.2] (0.01 1.1 1.0 lobeinancial problem? 2,941 (16.2] 1,530 (16.13] 1,411 (15.6} (0.05 0 9 (0.13?1.01 0.59 Eviction or loss of home 222(18] 31? 405 {0.01 1 3 1.4 School problem*? 162 (19.9] 20(118] 921121.19] N5 1 3 (0.9?1.81 1.3 Recent release from an institutioni"1 1,412 (16] 941 [10.2] (0.01 05 0.5 (04?05] JailfPrisonfDetention facility 203 (14.4] 82 121 (25.2] ?c001 3.6 [2.24.9] 4.5 HoSpital 513ll (36.6] 311 (33.0] 206 (43.7] (0.01 1 1.3 hospitaliinstitution 46933.2] 439146.21 30 <00] 01 0.1 Other [includes alcohDIISU treatment facilities} 2231153] 109 (11.6] 114 {24.2} 40.01 2 4 (I .8?33} 2.5 (1 3-3.3] See table footnotes on next page. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often ori? ented toward mental health conditions alone with regard to identification of suicidal persons, treatment of mental health conditions. and prevention titre-attempts. This study Found that approximately half of suicide. dcceclents in did not have a known mental health condition. indicating that additional Focus on nonmental health Factors Further upstream could provide important information For a public health approach (.10). Those without a known mental health condition suffered more From relationship problems and 622 MMWR June8,2018 Vol.6? No.22 other life stressors such as criminalr'legal matters. evicrionr?loss of home, and recent or impending criSes. Similarly, persons with mental health conditions also often experienced other circumstances such as relationship problems and johi'tinancial or physical health problems that contributed to their suicide. These findings point to the need to both prevent the circumstances associated with the onset of mental health conditions and support persons with known mental health conditions to decrease their risk for poor outcomes Two thirds ot'suicide decedents with mental health conditions had :1 history oi?treaonent for mental health or substance use disorders. US Department of Health and Human Servicestenters (or Disease Control and Prevention Morbidity and Mortality Weekly Report TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,ii 2015 Known No known mental health mental health Chi- Total condition+ condition square Adjusted 0R1 Characteristic No. No. No. p?value (95% Cl} [95% Cl} Crisis within past or upcoming 2 5,525 {29.41 2,444 {25.0} 3,081 {32.9} (0.01 1 4 1.411.345) Intimate partner problem 1.96.1356] 354134.51] 1,1141362} N5 1 1 1.1 Physical health problem 739 {13.41 315 {12.9) 424113.81 NS 1 1 [0.94.3] 1.0 Criminal legal problem 621 {11.2} 203 {8.31 41811345} (Bill 1.6 Family relationship problem 431] (18] 218111} {0.05 0.910.714.? Job problem 354 191 153 40.01 3? 0.7 {(15?03) Suicide eVentr?history Left a note 53581345] 3,1821333] 3,256 (35.1} NS 1.1 1.2 Disclosed suicide intent 4,405 (23.5] 2.0991214) corn 0.9 0.9 History of ideation 5.990 {31.9) 3,833 [40.8] 2,152 (23.0} I14 0.4 History of attempts 3,?32 {19.9} 2,?70 (29.41 962 {10.3) <0.01 [1.3 {0.3v0,3} 0.3 Abbreviations: ADDEADHD attention de?cit disorderrattention de?cit hyperactivity disorder; CI con?dence interval; WA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; St] substance use. Alaska. Arizona. Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan. Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Flhocle Island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or Iainr enforcement reports. i UR re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for sex. age group. and racer?ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. 1'1 The speci?c type of mental health condition was calculated only among those with one or more known diagnosed mental health conditions. 55 Not a diagnosis. 11?? Denominator is decedents aged 218 years. Denominator is decedents aged 10?1 3 years. 111 Denominator ofinstitution subgroup is decedents with recent release from an institution- Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroUp is decedents with any crisis within past or Upcoming 2 Weeks. Crises depicted here represent the most commonly occurring categories. with approximately half in treatment when they died. This find- ing suggests the [seed for additional safety supports, including broader implementation of affordable and effective treatment modalities, such as doctorrpaticnt collaborative care models and proven therapies. in addition, increased access to behavioral health providers in areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide preven- tion and patient safety, especially through care transitions Comprehensive sratcwidc suicide prevention activities are needed to address the full range offacturs contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household financial support): teaching coping and problem?solving slcills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and horror supporting persons at risk (cg, military veterans, persons with physical} mental health conditions) (12). Other Strategies include creating protective environments (cg, reducing access to lethal means among persons at rislc for suicide, creating organizational and US Department of Health and Human Servicesr?Centers for Disease Control and Prevention workplace policies to promote casing transitions into and our of work for persons with mental health conditions and other life challenges), strengthening access to and delivery of care, supporting family and friends alter a suicide. and cncourag ing the media to follow safe reporting recommendations (12). Some States, such as Colorado, are planning to implement such :1 comprehensive approach to suicide prevention The ?ndings in this report arc- to at least three limita- tions. First, in state-level analysis, rankings for four States (Maryland, Massachusetts, Rhodc island, and Utah) might have been affected by large proportions of injury deaths of undetermined [potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time {potentially biasing estimated ratc trends upward). Second, is not yet nationally representative; the 27 states included represent 49.6% ofthc population tmll. Finally, ahsrractors of data are limited to informa- tion contained in investigative reports. the extent of informant knowledge can data completeness and accuracy. Studies that include more in tcrvicws with nextaof-ltin often identify greater to mental health disorders MMWH June 8.2018 Vol.6? i No.22 511 Morbidity and Mortality Weekly Report Summary What is already known about this topic? In 2016. nearly 45,00!) pecans died by suicide in the United States. Mental health conditions are one ofseveralcontrihutorsto suicide. What is added by this report? During 999?201 a, suicide rates increased in nearly every state,- lncluding 80%. increases in 25 Staten 2015 data From 27 states indicate 54% of suicide decodents Were'not known tin-have mental health conditions. Relatitmship. substance use health..andjeb or ?nancial problems?a re among the other circumstances contributing to suicide. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as th use in PreventingSulcide: A Itchnical Package of Policy Programsand Practices, can help reach the national goal of reducing the annual suicide rate 20% by 2025. however. many methodological variations across studies exist it is likely that some persons without known mental health conditions in the current study were experiencing men? tal health challenges that were unknown, undiagnosed, or not reported by key informants. Nonetheless, the high preval of diverse contributing circumstances among those with and with out known mental health conditions suggests the importance of addressing the broad range offsetors that contribure to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are avail- able. States and communities can use data from and resources such as Suicide: A Echrricel Package off-bitty, Hogmms, and Practices [12) to better under- stand suicide in their populations. prioritize evidence?based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson. oily H'cdegaard. Margaret Warner. Division of Vital Statistics. National Center For Health Statistics. CDC. Con?ict of Interest No conflicts of interest were reported. IDivision of? 1Violence Prevention. National Center for injury Prevention and Control. CDC: 2Division of Analysis. Research. and Practice integration. National Center for lniury Prevention and Control, CDC. Corresponding author: Deborah M. Stone. Listonc?f?icdcgov. 624 MMWR June-8.2018 r' Vol.6? No.22 -I 9. ii]. References .CDC. Web?based lniury Statistics Query and Reporting System Atlanta, US Department of Health and Human Services, CDC. National Center for Injury Prevention and Control: 2?18. . [trey?Stephenson AZ, {:Irosby nE, Jack SPD. T, Sedacca Suicide trends among and within urbanization levels by sex. racelerhnicity. age group. and mechanism ofdeath?Unitcd States. 2001~2015. MMWR Surveill Surnm 2017;66lNo. 5548). i1ttps:lldoi. orgll U. Sal .Currin SC, Warner M. H. increase in suicide in the United States, 1999?3114. NCHS darn hricl?no. 24 l, Hyattsville, MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2016. db241.pdf . Kochnneic K. Murphy 5. Xu Arias E. Mortality in the United States, 2016. NCHS data brief-no. 295. Hyartwille, MD: US Department of Health and Human Services. (DC. National Center For Health Statistics; 2017. . Orlicc of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy For suicide prevention: goals and objectives For action. 1'ilif'sisl?iington. DC: US Department of Health and Human Services, (Ji?cc of the Surgeon General; 21]] 2. full-reportpdi . Zalsman C, Hawton K, D, ct Suicide prevention strategies revisited: 10?year review. Lancet K, O?Brien A. Leading suicide prevention cFForts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; .Crepcau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2illD:l?i:l4?34. 33l I 10905-47901 American Association. Diagnostic and statistical manual oi: mcnial disorders DC: American Association: 2013. Cainc ED. Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Prev 2017. Epub December .World Health Organization. Risks to mental health: an ovcrvicw oi vulnerabilities and risk factors. Geneva. Switzerland: World Health Organization: 2012. risks_ to_mental_hc3l .Stonc DM, Holland KM, Bartholow EN. Crosby All, Davis SP. Wilkins N. Preventing suicide: :1 technical package oipolicy. programs, and practice. Atlanta. US of Health and Human Services. 2017. suicidctechnicalpackagc.pdf . Milnerr?i. Sveticic J. De Leo D. Suicide in the absence disorder? A review of autopsy studies across countries. int] Soc 2015:59z?45?54. 1rtps:lldoi.orgfl {1.1 . Poulior Dc lco D. Critical issues in autopsy snidics. Suicide Lili: 'l'hrcar Belt-av US Department of Health and Human Serviceleenters for Disease Control and Prevention . Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report Weekly/Vol. 67 i No. 22 June 8, 2018 Vital Signs: Trends in State Suicide Rates United States, 1999?201 6 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Sch; Thomas R. Simon l?hDi; Katherine A. Fowler, Pth; Scott R. Kegler, Renting Yuan, Kristin M. Holland. Pth; Asha Z. hey-Stephenson, I?th; Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are one oF several factors contributing to suicide. Examining stateelevel trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia. Data From the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?20 16, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases Rates increased significantly among males and females in 34 and 43 states, rmpectively. Fifty-four percent ofdecedents in 27 states in 201 5 did not have a known mental health condition. Arnong decedents with available information, several circumstances were signi?cantly more likely among those without known mental health conditions than INSIDE among those with mental health conditions, including . relationship problems! loss (45.1% versus life 7 stressors (50.5% versus and recent! impending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and Without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidencerbased public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family members in the aftermath of a suicide. Continuing Education examination available at LLS. Department of Health and Human Services Centers for Disease Control and Prevention fsenor PK 5 N. I Morbidity and Mortality Weekly Report Introduction in 2016. nearly 45.000 suicides (15.6i100.000 population [age-adjusted? occurred in the United States among persons aged 2] 0 years From 1999 to 2015, suicide rates increased among both sexes. all racialiethnic groups, and all urbanization levels (2.3). Suicide rates have also increased among persons in all age groups <75 years, with ad - ged 45?64 having the largest posse-He increasii?l?S-Q??aefro [in per 100,000 persons ?1999llto 19.2 per 100,000 {gmo?and the greatesr number ofsuicides (232,108) during the same period Suicide is the 10th leading cause ofdeath and is one ofjust three leading causes that are increasing In addition. rates ofemergency department visits for nonfatal self-harm, a main risk factor for suicide, increased 42% from 2001 to 2016 (1). Together, suicides and selfdharm injuries cost the nation approximately $70 billion per year in direct medical and work loss costs (I). The National Strategy for Suicide Prevention calls for a public health approach to suicide prevention with efforts spanning multiple levels (individual, familyirelationship. community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite this call to action. suicide prevention largely focuses on identifying and referring suicidal persons to mental health treatment and pre- venting reattempts in addition to mental health conditions and prior suicide attempts, other contributing circumStances include secial and econornic problems. access to lethal means substances. firearms] among persons at risk. and poor coping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal. estab lished by the National Action Alliance of Suicide Prevention and the American Foundation for Suicide Prevention, of reduc- ing the annual suicide rate 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state?speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for strategies to include in comprehensive suicide prevention efforts that are based on the best available evidence. Methods Suicide rates were analyzed for persons aged :10 years because determining suicidal intent in younger children can be difficult Ageespecific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (interwoven! Classy?imtion of Diseases, 79ml: Revision. underlying?causc?ofdeath codes W370, U03). Age? population estimates were obtained from US. Census Bureaui?National Center for Health Statistics bridged-race population data releases. National and state?level suicide rate estimates were calr culated for six consecutive 3-year aggregate periods span- ning 1999?2016 (1999?2001; 2002?2004; 2005?2007; 2008?2010; 2011?2013; and 2014?2016). Rate estimates The Mm series of publications is published by the Centre: For Surveillance. Epidemioloy. and Laboratory Services. Centers for Disease Control and Prevention US. Department of Health and Human Services, ?tlant?a, GA 30329-4027. Suggested citation: [Amber names; ?rst three. then et aL. if more than six] [Report title]. MMWR Morb Mortal Wkly Rep 2018;67rlinclusive page numbers]. Centers for Disease Control and Prevention Robert R. Redfield. MD, Director Anne Schuciaat. MD. Principe! Deputy Dirrmr Les-lie Dauphin. Acting dentin-o Diwrmr?r Stir-Mr lemme Conn. MD. Dim-Mr. [wire @569an (23.!!in Chesley L. Richards, MD. MPH. De?ner Dimm?ir Pattie Haida Sniw?jic Serums Michael F. ladetnaico. MD. MPH. Din-tron Cenrsr?r Epidemial'a? and 11.2.50me Services MMWR Editorial and Produetion Staff {Weakly} Charlotte K. Kent. MPH. Acting Edirnrm Cliff: Executive Editor Jacqueline Gindler. MD. mar Mary Dott, MD, MPH. Undue Editor Tcresa F. Rutledge. Managing- Editor Douglas W. Weather-wait. Lead Tacenimi' Wm?Er?rar Glenn Damon. Soumya Dunwiarrh, Teresa M. Hood. MS. mantra! WrittrJEs?irarr Martha F. Boyd. Lead Visual In?rmatian Spec-?sh?s: Maureen A. Lea-shy. Julie C. Martinme. Stephen R. Spriggs. Tong Yang. Virtual In?rmetim swarm 'Quang M. Dean. MBA. H. King. Terraye M. Starr, Mona Yang, In?rmarian sweater. MMWR Editoriai Boats! Timothy F. Jones. MD. Chimera Matthew L. Boulton. MD. MPH Virginia A. Gains. MD Katherine Lyon Daniel. Jonathan E. Fielding, MPH, NIBA David Fleming. MD hJ MMWR i June-3.2018 r' Vol.6? I No. 22 William E. Helperin. MD. MPH King K. Holmes. MD. Robin lkeda. MD. MPH Rims F. Khahbaa. MD Meadows. MSN. RN Jewel Mullen. MD, MPH. MFA jeEfNiedet-deppe. rants; omits. MD. MPH Patrick L. Remington..MD. MPH Curios Roig. MS. MA William L. Roper. MD. MPH William Scha?iner. MD US Department of Health and Human Servicesr'Cenlers for Disease Control and Prevention PROOF PROOF PROOF PROOF Morbidity and Mortality Weekly Report were age-adjusted to the US. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted leasr?squates regression with inverse?variance weighting. Modeled rate trends are reported in 01: average an [1.1131 Chang?'s. Characteristics of persons aged 3:10 years who died by suicide, with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 27 states? with complete data participating in National Violent Death Reporting System in 2015. defines mental health conditions as disorders and lisred in the Diagnostic and Statistical Man as! of'Mmriri' Disorders, Iii/in edition with the exception of problematic alcohol use and other substance use that are cap- tured Separately in aggregates data From three primary data sources: death certificates, coroner!I medical examiner reports (including toxicology), and law enforcement reports. A range ofcircumstanccs (relationship problems, li?: stressors. and recent or impending crises] have been identified as potential risk Factors For suicide in Circumstances captured are those identified as contributing to suicide in coronerimedical examiner or law enforcement reports, which re?ect information provided by Family and Friends at the time of death. Decedenrs could have experienced multiple circum- stances. Decedents with and without known mental health conditions were compared using chi?Square tests. Logistic regression analyses were used to estimate adjusted odds ratios (aURs) with 95% con?dence intervals (C15), controlling for sex, age group, and raceiethniciry. Results The most recent overall suicide rates (representing 2014? 2016} varied fourfold, From 6.9 {District of Columbia) to 29.2 {Montana} per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all srates except Nevada (where the rate was consistently high throughout the study period], with absolute increases ranging From 0.8 per 100,000 {Delaware} to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases 330% observed in 25 states (Supplementary Table; (Figure). Modeled suicide rate trends indicated significant increases in 44 states, among males (34 states) and Females (435 states), as well as for the United States overall (Supplementary Table; Alaska, Arizona, Colorado, Connecticut. lfieorgia, Hawaii, Kansas, Kentucky, Maine. Maryland, Massachusetts. Michigan, Minnesota, New Hampshire, New jersey, New Mexico, 1.?slew York. North Carolina, Ohio, Oklahoma, Oregon, Rhode Island. South Carolina, Utah, Vermont, Virginia. and Wisconsin. US Department of Health and Human Services/Centers for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 . I Increasez3B%?SB% I ncrease119%?30% Increase: ass-13% 1:1 Decrease: Heat: Per 100,000 population, age?adjusted to the 2000 standard population. Nationally, the model?estimated average annual percentage change For the overall suicide rate was an increase of1.5%. By sex, estimated national rate trends Further indicated significant average annual percentage change increases For males and Females (Supplementary Table; rdci537851. Suicide decedents without known mental health conditions (1 1,039; 54.0%} were compared with those with known men? tal health conditions (9,407: 46.0%) for 27 states. Whereas decedents were predominantly male (Table 1) and non?Hispanic white those without known mental health conditions, relative to those with mental health con? ditions, were more likely to be male (83.6% versus 63.8%; odds ratio 2.3, 95% C1 2.2?2.5) and belong to a racialiethnic minority (OR range Suicide decev dents without known mental health conditions also had signi?cantly higher odds of perpetrating homicide Followed by suicide 2.9, 95% CI Among decedents aged 218 years, 20.1% ofthose without known mental health conditions and 15.3% ofthose with mental health conditions had previously served in the 1.1.5. military or were serving at the time oF death. Whereas Firearms were the most common method oFsuicide overall decedents without known mental health conditions were more likely to die by firearm and less likely to die by or poisoning than were those with known mental heal Lh MMWR i" June 8,2018 i Vol.6? i No.22 3 PROOF PROOF PROOF PRUOF Morbidity and Mortality Weekly Report TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states..* 2015 EL Known mental No known mental ealth conditionf health condition Chi?square Adjusted I001l Characteristic EN (n (n 1 p-VaIUe {95% Cl] [95% Cl] Sex Male 15.702 (26.3] 6.469 (08.3) 9.233 (33.5) c0.0l 2.3 NA Female 4.744 (23.2] 2.938 (31.2) 1.806 (16.4) {0.01 NA Ase airs)? 10-24 2,304 (13.2] 1.211 (12.9) 1.593 (14.4) (0.0) 1.1 NA 25?44 6,456 (31.6] 3,036 (32.3) 3.420 (31.0) {0.05 0.9 NA 45?64 7.71 {37.71 3.820 (40.6} 3.398 (35.3i <0.0'i 0.3 NA 265 3.463 {17.01 1.340 (14.2) 2.123 (19.3) 1.4 0.3?1.5} NA RacefEthnicity White. non~Hi5panic 12.102 {33.61 (86.8) 3.93? (31.0) ?0.01 0.6 NA Black. non~Hispanic 1.228010] 411 81? (7.4) {0.01 NA American Indiani?AIaska Native. are 112 266 <00} 2.0 (l 15?215) NA non?Hispanic Asian. non-Hispanic SIG 235 (2.5) 341 (3.1) eons 1.2 0.1-1.5) NA Hispanic 1.096 46:31:41.9} 533 (5.7) {0.05 1.2 (1.0-1.3) NA Other 65 21 (0.2) 45 (0.4) <0.05 NA Extended demographics Ever served in militaryiJr 3.429 (17.3) 1.354 (15.3) 2.075 (20.1) ??0.01 1.4(1 3?1.5) 1.1 Homeless 24-13(12) 104 (1.1) 136 (1.3) NS 1.1 1.2 Incident type Single suicide 20.053 (98.2] 9.313 (99.1) 10.745 (92.4) <0.01 0.3 0.4 Homicide followed by suicide 319(1.6] 6410?} 255 (2.3) {0.01 3.5 2.9 Multiple suicides 64 25 (0.3) 39 (0.4) NS 1.3 (0.8-2.2) 1.5 (0.9-2.6) Method Firearm 9,909 (48.5] 3.321 (40.0) 5.038 (55.3) (0.01 1.8 1.0 5.907 (23.9] 2.940 (31.3) 2.967 (26.9) (0.01 0.8 (0.8-0.9) 0.8 Poisoning 3.003 (14.2] 1,361 (19.3) 1.142 (10.4) {0.01 0.5 0.6 Substance class causing cleat?nl?? Other over?thE-counterl 1.021 {34.01 656 (35.8} 355 (31.1) 0.8 (0.2-0.9) 0.9 (0.2-1.0) Dpioids 944 (31.4] 608 (32.7] 336 (29.4) NS 0.9 (0.24.0) 0.9 Antidepressants 800 {20.5) 644 (34.6) 156 (0.01 0.3 0.3 (0.3?0.4l Be nzodlazepines 624 {20.8} 458 {25.1} 156 {13.7) I15 {DA?tin) 0.5 219 195 (10.5) 24 (2.1) {0.01 0.2 0.2 (0.1-0.3) Other 1,595 2'80 (8.3) 315 (7.4) 4.0.05 0.9 0.9 See table footnotes on next page. conditions 31.3%, and 19.8%. respectively]. These remained significant in the adjusted models. Toxicology testing was less likely to be performed liar decer dents without known mental health conditions. Among those with toxicology results. decedents without known mental health conditions were less likely to test positive for any subsrance overall 0.8. 95% Cl including opioids 0.90. 95% CI but were more likely to test positive for alcohol 1.2. 95%, CI Information on circumstances surrounding suicide were available for all decedents with mental health conditions (9.407) and approximately 85% of those without known mental health conditions (9.5557) in 2? states (Table 3). Persons without known mental health conditions were less likely to have any problematic substance use 0.7. 95% Ci 0.7?0.3) than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health 1 MMWR June8.2018 r' Vol.6? 1 No.22 or substance use treatment just over half were in treatment at the time oideath. Decedents without known mental health conditions had a significantly higher likelihood of any relationship problem.ll loss than did those with known mental health condi- tions speci?cally intimate partner problems (30.2% versus argumentsiconflicrs (17.5% versus and perpetrating interpersonal violence in the past month versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (50.5% versus 47.2%) such mil-criminal legal problems (10.7% versus or eviction/loss ofhome versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks. respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained sig- nificant in the adjusted models. Physical health problems and US Department of Health and Human Servicestenters for DiseaSE Control and Prevention PROOF PROOF Morbidity and Mortality Weekly Report Paoor PRL or TABLE 1.1Conrr?nued] Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health conditiorIJr health condition Chi-square 011E Adjusted 0H1 Characteristic in {n 1 1,03% p?value {95% (95% Cl} Toxicology results Any toxicology testing 13,31? {55.1} 5,658 {70.8} 5,559 {60.3} (0.01 0.6 0.7 Positive for at substance?l 9,913 [14.4) 5,192 {1?80} 4,?21 {10.9} (0.01 0.5 Substance detected?i? Alcohol Tested 10,950 [53.6) 5,409 {52.5} 5,541 {50.2} (0.01 0.7 0.3 Positive 4,4421400} 2,115 (39.1} 2,32? {42.0} (0.01 1.1 1.2 Opioids Tested 8,554 141.8} 4,253 (45.3} 4,296 {33.9} (0.01 0.8 0.8 Positive 2,229 {26.6} 1,238 (29.1} 1,041 {24.2} (0.01 0.8 0.9 Benzodiazepines Tested 3,124 {39.2} 4,226 {44.9} 3,393 (35.3} (0.01 0.7 Positive 2.4641303} 1,639 {33.3} 325 {21.2} <0.01 0.4 0.5 Cocaine Tested 2,973 {39.0} 3,365 041.1} 4,112 {32.2} (0.01 0.9 (0.8?0.91 0.9 i0.9?1.0} Positive 49916.3} 216 233 (6.91 (0.05 1211.04.51 1.2 {1.0?1.51 Amphetamines Tested 2,615 {32.2} 3,596 {39.3} 3,919 135.5} <0.01 0.9 (0.13-0.91 0.9 Positive 136 316 {10.2} 350 [9.21 NS 0.9 1.0 {0.3?1 Marijuana Tested 6,569 {32.1} 3,127 (33.2} 3,442 {31 {0.01 0.9 0.9 [0.94 Pasitlve 1,421 {22.4} 210122.71 1'61 {22.1} NS 1.0 (0.9?1 0.9 Antidepressants Tested 5,425 {26.5} 3,103 {33.0} 2,322 121.0} {0.01 0.5 0.6 Positive 23141403} 1,?35 155.9} 479120.51 (0.01 0.2 0.2 Abbreviations: CI confidence interval; NA 2 not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma. Oregon. Rhode Island, South Carolina, Utah, Vermont. Virginia. and Wisconsin. Decedent had been identi?ed as having a current diagnosis of mental health condition In coronerr'medical examiner or law enforcement reports. 5 DR reflects the risk among those without known mental health condition relative to those with knoWn mental health condition. '1 Logistic regression was used to estimate adjusted DP. with 95% after controlling for sex, age group. and racefethnicity. Known mental health condition was used as the reference group. ?i Decedents were aged 210 years. as per standard in the suicide prevention literature. Denominator is decedents aged 218 years with reported military service status. it Denominator is decedents who died by poisoning, including overdose. Denominator is decedents with any toxicology testing. Denominatorfor each positive group is the number tested for the substance in that group. joblfinancial problems commonly contributing stresso rs among both persons without mental health conditions (23.2% and 15.6%, respectively} and those with mental health condi? tions (21.4% and 16.8%, respectively}. Similarly, among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. without known mental health conditions had significantly lower odds of recent release from any institu- tion 0.5. 95% CI Amang those recently rclcaScd, decedents without known mental health conditions were signi?cantly more likely than with mcntai health conditions to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other Facility, such as an alcohoilsubstancc use treatment facility [24.2% versus 11 .60/01.Among decedents with known mental health conditions who were recently released fro an institution, 46.7% were released from facilities. US Department of Health and Human Servicesr?Centers for Disease Control and Prevention without known mental health conditions were significantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%. respectively}. Suicide intent was disclosed by 22.4% and 24.5% ofpersons without and with known mental health conditions. respectively. Conclusions and Comments During 1999?2016, suicide rates increased in 44 states, and 25 states experienced increases 260%. Rates increased significantly among males in 34 states, and females in 43 states. Additional research into the specific causes of these trends is needed. Data from the 27 states participating in provide important insight into cirr cumstances surrounding suicide and can help states identify prevention priorities. MMWR June 8. 2018 Vol.6? r' No. 22 f1 PROOF Paoor Pnoor Paoor Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged 1:10 years with and without known mental health conditions National Violent Death Reporting System, states,* 2015 Known No known mental health mental health Chir condition?i condition square 0115 Adjusted 0111 Characteristic Total no. no. p-ualue 195% C11 195% Suicide with known circumstances 18.264 [91.81 9.407 11001 9.357 134.31 6.0.01 NIA Mental health Any current diagnosed mental health condition? ?1'i 2,076 125.21 NRA MA MA MA Anxiety disorder ?t1 1.5?911531 WA MA MA Bipolar disorder +1 1.431 115.21 ma WA Nra WA Schizophrenia 509 15.4] MIA MA MA PTSD ?11 424 14.51 NM MA MA NHA 1* 22612.41 MIA WA MA Not specified #11 1'60 18.1] are Current depressed 111:dei? 7,033 (31.51 335214111 3,015 132.91 {0.01 or or (as?arr Problematic substance use Any 5,319 123.31 2.976 131.6] 2.343 125.01 <0.01 0.2 0.2 Alcohol 3.2681114} 1.862 119.81 1,406 115.01 <0.01 Other 3,084 [16.41 126811331 1.3161141} <0.01 01103?031 13.210.741.81 Treatment Current mental healthr'substance use treatment 5.141 122.41 5,077 154.01 64 10.21 <0.01 0.01 10.01~0.011 0.01 10.01 ?0.011 Eyer treated for mental healt hrsuhstance disorder 6.71 2 [35.81 6,323 162.21 394 14.21 {0.01 0.02 {0.02?0.02} 0.02 {0.02?0.03} Relationship problemsiloss Any relationship problemr'loss 194314241 122613961 4,222 145.11 <0.01311 1 3.1 1.3 Intimate partner problem 5.0901212} 2.2201241] 2,323 130.21 {0.01 141.3?1 .51 1.4 Perpetrator ofinterpersonal violence in past month 41412.2) 131 11.41 283 13.01 (0.01211 B.- 2 .21 2011.64.41 lliictim of Interpersonal violence in past month 3410.41 5310.61 31 10.31 <0.05 610. 4?0. 91 0.0 {0.5?1 .21 Family relationship problem 1.621 13.91 323 19.3] 295 (3.1.0 {0.9?1 .11 Other relationship problem lnonintimatel 40312.11 20212.1] 201 12.11 NS 1.0 1.1 Argument orcon?ict lnot specified) 2.9141155} 1.278 113.61 1.636 117.51 c001 1311.24.51 1.4 Death ofa loved one {any} 1,49? [3.01 326 {3.31 621 17.21 <0.01 0.3 10.71?11.91 0.9 Nonsuicide death 1,181 [6.31 541' {6.91 53415.21 c001 0.3 107?091 0.53 [0.11?1.01 Suicide of family or friend 37912.01 21 3" 12.31 162 11.71 (0.01 0.2 10.6e0_91 0.8 (0.71-1.01 Other life stressors Any life stressor 9.171 140.91 4.442 142.21 4.7291505) 6.0.01 1.1 1.1 Recent criminal legal problem 158818.51 58616.21 1,002 110.21 (0.01 1.811.6?201 Other legal problem 745 [4.01 3213 {4.01 320 14.01 NS 1.0 {0.8?1 1.0 Physical health problem 4.129 122.31 2,012121.41 2.167 123.21 c001 1 11.04.21 1011.04.11 Jobrnnanciel problem? 2.941 115.2) 1,530 (15.31 1,411 {15.51 was as (as-1.01 13.910.34.01 Eviction orloss ofhome 72213.01 31213.41 40514.31 <0.01 1.3 1411.24.61 School problem*? 162119.91 2'0 {12.131 92121.9} NS [0.9?1.3113103?131 Recent release from an institution?rtt 1.412 on} 941 110.21 471 15.11 ?3.01 510. 4-11. 51 as roe-0.51 JailfPrisonfDetention facility 203 [14.41 8218.21 121 {25.21 (0.01612. 24 914.5 Hospital 512136.61 311 {33.01 206143.21 <0.01611 2 01 1.3 hospitalr'institution 469133.21 439146.71 3016.41 <0.01 0.1 0.1 {0.1?0.11 Other {includes alcoholrSU treatment facilities} 223115.81 109111.61 114 {24.21 <00? 14113-33} See table footnotes on next page. Suicidologists regularly state that suicide is not caused by a single factor (51; however, suicide prevention is often oriv cnted toward mental health conditions alone with regard to identi?cation of suicidal persons. treatment of mental health conditions, and prevention This stud},r found that approximately half of suicide in did not have a known mental health condition. indicating that additional Focus on health factors further upstream could provide important information For a public health approach Those without a known mental health condition suffered more from relationship problems and 6 MMWR Jone8.2018 r' Vol.6? 1 No.22 other life stressors such as criminal/legal matters, evictionr'loss ofhome, and recent or impending crises. Similarly, persons with mental health conditions also often experienced other circumstances such as relationship problems and jobr'tinancial or physical health problems that contributed to their suicide. These findings point to the need to both prevent the circumstances associated with the onset of mental health conditions and support persons wid'i known mental health conditions to decrease their risk For poor outcomcs (.1 Two thirds ofsuicidc with mental health conditions had a history of treatment For mental health or substance use US Department of Health and Human Sewicesr?Centers for DiseaSE Control and Prevention PROOF FWMDOF Morbidity and Mortality Weekly Report PROOF TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System. 27 states,?i 2015 Known No known mental health mental health Chi- condition+ condition square (21115 Adjusted Characteristic Total no. no. [as] p?value {95% {95% Cl} Crisis within past or upcoming 2 5,525 {29.4} 2,444 {25.0} 3,031 {32.9} mm 1.4 1.4 {1 .3?1 .5) Intimate partner problem 1,968 {35.6} 854 {34.9} 1.114 {36.2} NS 1.1 1 .1 Physical health problem 1?39 {13.4} 315 {12.9} 424 {13.8} NS 1.1 1.0 Criminal legal problem 621 {11.2} 203 418 {13.6} (0.01 1.6 Family relationship problem 4311 {13} 212 213 {0.05 0.8 0.9 Job problem 354 191 163 (0.01 Suicide eventfhistory Left a note 6.4681345} 3,132 {33. 3} 3.286 {35.Disclosed suicide intent 4,405 {23.5} 2.306 {24. 5} 2.1199 {22.4} (0.131 0.9 t} 9 it} B- 9} History of ldeation 5,991.1 {31.9} 2,152 {23.11} 30?10. Rates increased significantly among males in 34 states. and females in 43 states. Additional research into the specific causes of these trends is needed. Data from the 27 states participating in provide important insight into cirr cumstanccs surrounding suicide and can help states identify prevention priorities. MMWR June 8. 2018 Vol.6? it No. 22 5 PROOF Paoor Pnoor Paoor Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged 1:10 years with and without known mental health conditions National Violent Death Reporting System, states,* 2015 Known No known mental health mental health Chir condition?i condition square 0115 Adjusted 0111 Characteristic Total no. no. p-ualue 195% C11 195% Suicide with known circumstances 18.264 [91.81 9.407 11001 9.357 134.31 6.0.01 NIA Mental health Any current diagnosed mental health condition? ?1'i 2,076 125.21 NRA MA MA MA Anxiety disorder ?t1 1.5?911531 WA MA MA Bipolar disorder +1 1.431 115.21 ma WA Nra WA Schizophrenia 509 15.4] MIA MA MA PTSD ?11 424 14.51 NM MA MA NHA 1* 22612.41 MIA WA MA Not specified #11 1'60 18.1] are Current depressed 111:dei? 7,033 (31.51 335214111 3,015 132.91 {0.01 or or (as?arr Problematic substance use Any 5,319 123.31 2.976 131.6] 2.343 125.01 <0.01 0.2 0.2 Alcohol 3.2681114} 1.862 119.81 1,406 115.01 <0.01 Other 3,084 [16.41 126811331 1.3161141} <0.01 01103?031 13.210.741.81 Treatment Current mental healthr'substance use treatment 5.141 122.41 5,077 154.01 64 10.21 <0.01 0.01 10.01~0.011 0.01 10.01 ?0.011 Eyer treated for mental healt hrsuhstance disorder 6.71 2 [35.81 6,323 162.21 394 14.21 {0.01 0.02 {0.02?0.02} 0.02 {0.02?0.03} Relationship problemsiloss Any relationship problemr'loss 194314241 122613961 4,222 145.11 <0.01311 1 3.1 1.3 Intimate partner problem 5.0901212} 2.2201241] 2,323 130.21 {0.01 141.3?1 .51 1.4 Perpetrator ofinterpersonal violence in past month 41412.2) 131 11.41 283 13.01 (0.01211 B.- 2 .21 2011.64.41 lliictim of Interpersonal violence in past month 3410.41 5310.61 31 10.31 <0.05 610. 4?0. 91 0.0 {0.5?1 .21 Family relationship problem 1.621 13.91 323 19.3] 295 (3.1.0 {0.9?1 .11 Other relationship problem lnonintimatel 40312.11 20212.1] 201 12.11 NS 1.0 1.1 Argument orcon?ict lnot specified) 2.9141155} 1.278 113.61 1.636 117.51 c001 1311.24.51 1.4 Death ofa loved one {any} 1,49? [3.01 326 {3.31 621 17.21 <0.01 0.3 10.71?11.91 0.9 Nonsuicide death 1,181 [6.31 541' {6.91 53415.21 c001 0.3 107?091 0.53 [0.11?1.01 Suicide of family or friend 37912.01 21 3" 12.31 162 11.71 (0.01 0.2 10.6e0_91 0.8 (0.71-1.01 Other life stressors Any life stressor 9.171 140.91 4.442 142.21 4.7291505) 6.0.01 1.1 1.1 Recent criminal legal problem 158818.51 58616.21 1,002 110.21 (0.01 1.811.6?201 Other legal problem 745 [4.01 3213 {4.01 320 14.01 NS 1.0 {0.8?1 1.0 Physical health problem 4.129 122.31 2,012121.41 2.167 123.21 c001 1 11.04.21 1011.04.11 Jobrnnanciel problem? 2.941 115.2) 1,530 (15.31 1,411 {15.51 was as (as-1.01 13.910.34.01 Eviction orloss ofhome 72213.01 31213.41 40514.31 <0.01 1.3 1411.24.61 School problem*? 162119.91 2'0 {12.131 92121.9} NS [0.9?1.3113103?131 Recent release from an institution?rtt 1.412 on} 941 110.21 471 15.11 ?3.01 510. 4-11. 51 as roe-0.51 JailfPrisonfDetention facility 203 [14.41 8218.21 121 {25.21 (0.01612. 24 914.5 Hospital 512136.61 311 {33.01 206143.21 <0.01611 2 01 1.3 hospitalr'institution 469133.21 439146.71 3016.41 <0.01 0.1 0.1 {0.1?0.11 Other {includes alcoholrSU treatment facilities} 223115.81 109111.61 114 {24.21 <00? 14113-33} See table footnotes on next page. Suicidologists regularly state that suicide is not caused by a single factor (51; however, suicide prevention is often oriv cnted toward mental health conditions alone with regard to identi?cation of suicidal persons. treatment of mental health conditions, and prevention This stud},r found that approximately half of suicide in did not have a known mental health condition. indicating that additional Focus on health factors further upstream could provide important information For a public health approach Those without a known mental health condition suffered more from relationship problems and 6 MMWR Jone8.2018 r' Vol.6? 1 No.22 other life stressors such as criminal/legal matters, evictionr'loss ofhome, and recent or impending crises. Similarly, persons with mental health conditions also often experienced other circumstances such as relationship problems and jobr'tinancial or physical health problems that contributed to their suicide. These findings point to the need to both prevent the circumstances associated with the onset of mental health conditions and support persons wid'i known mental health conditions to decrease their risk For poor outcomcs (.1 Two thirds ofsuicidc with mental health conditions had a history of treatment For mental health or substance use US Department of Health and Human Sewicesr?Centers for DiseaSE Control and Prevention PROOF PROOF Morbidity and Mortality Weekly Report HOG TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System. 27 states,? 2015 Known No known mental health mental health Chi- condition+ condition square 0R5 Adjusted 'l.'.1Fi'n Characteristic Total no. no. [ii-'13} p?value (95% Cl} [95% Cl} Crisis within past or upcoming 2 5.525 {29.4} 2.444 {25.0} 3.081 {32.9} <03} 1 4 Intimate partner problem 1,968 {35.6} 854 {34.9} 1.114 {36.2} N5 1 1 1.1 Physical health problem 1?39 {13.4} 315 {12.9} 424 (13.3} NS 1 1 1.0 Criminal legal problem 621 [11.2) 203 418 {13.6} (0.01 1.6 Family relationship problem 431.} [13} 212 218 {11} ?(0.05 8 0.9 [03?1 Job problem 35416.4} 191 [18} 153 3? 0.7 {(15?03) Suicide evantfhistory Left a note 6.463 (34.5} 3.132 (33.3] 3.236 {35.1} NS 1.1 1.2 Disclosed suicide intent 4.4115 123.5} 2.306 (24.5} 2.099 {22.4} mm 0.9 0.9 History of ideation 5.990 (31.9} 3.838 (49.8} 2.152 0.4 History of attempts 3.732 [19.91 2.370 {29.4} 962 {10.3} 40.01 [1.3 0.3 {(13?03} Abbreviations: ADDEADHD attention de?cit disorderiattention deficit hyperactivity disorder; Cl 2 con?dence interval; MIA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; St] substance use. Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma, Oregon. Rhode island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. DR re?ects the rislr among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for sex. age group. and raceiethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions. which are not mutually exclUsive. Therefore. sums of percentages for the diagnoSed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. The specific type of mental health condition was calculated only among those with one or more known diagnosed mental health conditions. 95 Not a diagnosis. 11?? Denominator is decedents aged 218 years. Denominator is decedents aged 10?1 8 years. 111 Denominator ofinstitution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. 5% Denominator of crisis subgroUp is decedents with any crisis within past or upcoming 2 Weeks. Crises depicted here represent the most commonly occurring categories. disorders. with approximately half in treatment when they died. This finding suggests the need for additional safety supports. including broader implementation ofaffordablc and effective treatment modalities. such as doctorapaticnt collaborativc care models and proven cognitive?behavioral therapies. in addition. increased access to behavioral health providers in areas is needed. as is expansion of health care systems that integrate physical and behavioral health. with a priority on suicide prevention and patient safety. especially through care transitions (12). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports (cg. housing stabilization policies. household ?nancial support); teaching coping and problem?solving skills to manage everyday strossors and prevent future relationship problems. especially early in life; promoting social connectedness to increase 3 sense of belonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk leg. military veterans. persons with physicalf mental health conditions) Other strategies include creating protective environments (cg. reducing access to lethal means US Department of Health and Human Services/Centers for Disease Control and Prevention among pert-ions at risk for suicide. creating organizational and workplace policies to promote help-seeking. casing transitions into and out of work for persons with mental health conditions and other life challenges}. strengthening access to and delivery of care. supporting family and friends after a suicide. and assuring the media. follow safe reporting recommendations U2). Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention [170). The findings in this report are subject to at least three limiter rions. First. in the state-level analysis. rankings for four states (Maryland. Massachusetts. Rhoda: island. and Utah) might have. been affected by large proportions of injury dearhs of undetermined intent (potentially biasing reported suicide rates downward] or decreased rages of such deaths over time (potentially biasing estimated rate trends upward}. Second. is not yet nationally representative; the 2? states included 49.6% of the population Finally. abstractors of data are limited to information contained in investigative reports. Therefore. the extent ofinformant knowledge can affect data completeness and accuracy. Studies that include more interviews MMWH a" June 8. 2018 Vol.6? No.22 'Mi PROOF PROOF ROOF PROOF Morbidity and Mortality Weekly Report Summary What is already known about this tragic? In an 6, nearly 45,0tiupersons died by suicide in the United States. Mental health conditions are one suicide. What is added by this renort? During 1999?2615. suicide rates increased in nearly every state,- including )30% increases irr25 data From 27 states indicate 54% of suicide decedents were not .ltr?rovvr?r to have tnental health conditions. Relationship, substance use, health, and job or ?nancial problems are among the other circumstances contributing to suicide. What are the rut pllcations for public health practice? A-comprehemive approach using proven prevention strategies, such asthosein Technical Package for Suicide Prevention, can help readt the national goal of reducing the annual suicide rate 20% by 2025. with next?of?ltin often identify greater attributions to mental disorders however, many methodological variations across studies exist it is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown, and hence underreported by key in to rrnants. Nonetheless, the high prevalence ofdiverse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range oF factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are avail- able. States and communities can use data From and resources such as Preventing Suicide: fl Ethanol Package quar'icjr Programs. and Practices [12) to better under? stand suicide in their populations, prioritize evidenceebased comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard. Margaret Warner, Division of Vital Statistics, National Center for Health Statistics, CDC. Con?ict of Interest No con?icts of interest were reported. lDivision of Violence Prevention. National Center for injury Prevention and Control, zDivislnn of? Analysis. Research, and Practice Integration, National Center For Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone. El i .lLllieB. 2018 Vol.6? i No. 22 6. 9. ill. l4. References .CDC. Web?based injury Statistics Query and Reporting System ntlanta, CA: National Center for Injury Prevention and Control: 2U18. . hey?Stephenson AZ. Crosby RE, Jack Hailcyesus T. Sedacca Suicide trends among and within urbanization levels by sex, age group, and mechanism States, 2001?2Ul5. MMWR Surveill Summ . Curtin SC. Warner M. l'ledegaard H. Increase in suicide in the United States, l999?2l?r14. NCHS data hricf no 241. Hyartsville. MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. db24l.pdf . Koclranelt K, Murphy 5, Xu J, Arias E. Mortality in the United States, 20] NCHS data brief no 2.93. Hyattsville, MD: US Department of Health and Human Services. CDC. National Center For Health Statistics; 201?. . Office of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives For action. \Vashington, DC: US Department of Health and Human Services, Cll?lice oi: the Surgeon General: 2012. ibraryl tionl Full-report.de Zalsman G, Hawron K, Wasserman D, er al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 20 D. [0 1 61?5le 5?0366l .Tor'guson K, O?Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation For Suicide Prevention; ZUIT. .Crepcau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 1 American Association. Diagnostic and statistical manual of mental disorders Washington. DC: American Association; 20H. Caine ED. Reed J, Hindmarr J. Quinlan K. Comprehensive, integrated approaches to suicide prevention: pracrical guidance. Prev 201?. Epub December 20, 2131?. l??r?injuryprev?le 17?042366 .World Health Organization. Risks to mental health: an overview of vulnerabilities and risk Factors. Geneva. Swirrerland: World Health Organization: 2012. Stone DM. Holland KM. Bartholow BN. Crosby AE, Davis SP. Wilkins N. Preventing suicide: 3 technical package ol?policy, programs, and practice. Atlanta. GA: US Department of Health and Human Services, CDC: 20W. suicidetechnicalpackagepdf . Milncr A. Svcticic J, De Lco D. Suicide in absence ol?mental disorder? A revicvrr of autopsy studies across countries. Soc 0.1 Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 2rrss;3s;4sr?sr 0. US Department of Health and Human Servicestenters for Disease Control and Prevention . Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report Weekly/Vol. 67 i No. 22 June 8, 2018 Vital Signs: Trends in State Suicide Rates United States, 1999?201 6 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Sch; Thomas R. Simon l?hDi; Katherine A. Fowler, Pth; Scott R. Kegler, Renting Yuan, Kristin M. Holland. Pth; Asha Z. hey-Stephenson, I?th; Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are one oF several factors contributing to suicide. Examining stateelevel trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia. Data From the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?20 16, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases Rates increased significantly among males and females in 34 and 43 states, rmpectively. Fifty-four percent ofdecedents in 27 states in 201 5 did not have a known mental health condition. Arnong decedents with available information, several circumstances were signi?cantly more likely among those without known mental health conditions than INSIDE among those with mental health conditions, including . relationship problems! loss (45.1% versus life 7 stressors (50.5% versus and recent! impending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and Without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidencerbased public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family members in the aftermath of a suicide. Continuing Education examination available at LLS. Department of Health and Human Services Centers for Disease Control and Prevention PREHDF Petite Morbidity and Mortality Weekly Report senor 2>aotnr Introduction in 2016. nearly 45,000 suicides (1 population [age-adjusted? occurred in the United States among persons aged :10 years [i i. From 1999 to 2013, suicide rates increased among both sexes, all racial/ ethnic groups, and all urbanization levels Suicide rates have also increased among persons in all age groups <75 years, with adults aged 45?64 having the largest absolute rate increase {From 13.2 per 100,000 persons [1999] to 19.2 per 100,000 [2016]) and the greatesr number oi?suicides (232,108) during the same period (1.3). Suicide is the 10th leading cause ofdeath and is one oi'just three leading causes that are increasing in addition, rates ofeinergency department visits for nonfatal self-harm, a main risk Factor For suicide, increased 42% From 2001 to 2016 Together, suicides and Self-harm injuries cost the nation $70 billion per year in direct medical and work loss costs if). The National Strategy For Suicide Prevention calls for a public health approach to suicide prevention with efforts span? ning multiple levels (individual, Familyfrelationship, community, and societal]. Such a comprehensive approach underscores that suicide is rarely caused by any single Factor, but rather, is deter? mined by multiple Factors. Despite this call to action, suicide prevention largely focuses on identifying and referring suicidal persons to mental health treatment and preventing reattempts (6). In addition to mental health conditions and prior suicide attempts, other contriburing circumsrances include social and economic problems, access to lethal means (cg, substances, ?rearms) among persons at risk, and poor coping and problem? solving skills Expanded awareness of these additional circum- stances contributing to suicide risk and action to address them can help reach the national goal, established by the National Action Alliance of Suicide Prevention and the American Foundation For Suicide Prevention, of reducing the annual suicide rate 20% by 2025 (7). To assiSt states in achieving this goal, CDC analyzed state?specific trends in suicide rates and assessed the multiple contributing Factors to suicide; this report presents options For Strategies to include in comprehensive suicide prevention efiorts that are based on the best available evidence. Methods Suicide rates were analyzed for persons aged :10 years because determining suicidal intent in younger children can be difficult (8). Ageaspecil'ic suicide counts were tabulated based on National Vital Statistics System coded death certificate records (Internnrianoi Closs?imiion afDiseoses, ?nch Revision, underlying?cause?ofdeath codes ?37.0, Age? specific popuiation estimates were obtained From U.S. Census Bureau/National Center For Health Statistics bridged?race population data releases. National and staterlevel suicide rate estimates were calculated for six consecutive 3?year aggregate periods spanning 1999? 2016 (1999?2001; 2002?2004; 2005?2007; 2003?2010; 2011?2013; and 2014-2016). Rate estimates were age- adjusted to the U.S. 2000 standard population and expressed The of publications is published by the Center For Surveillance. Epidemiology, and Laboratory Services. Centers For Disease Comm] and Prevention (CDC). U.S. Department of Health and Human Sentices, ?tlant?a, GA 30329-4027. Suggested cim?n'ne [Audaor tuna-2;; ?rst three. then et aL, ifmore than sire} [Rt-port title]. Mm Morb Mortal Wkly Rep 2018;67:linc1usiva page numbers]. Centers for?Disease Control and Prevention Robert R. Redfield. MD, Director Anne Sehuehat. MD. J?s-impel Depend Dimmi- Les-lie Dauphin, Acting Annrim Dirermr?r Sol-nee Joanne Conn. MD. Dim-ran [wire ofScimi-e Qmig'gr Chesley L. Richards, MD. MPH. Doom}! Pu Mir Health Srioai'y?ir Services Michael F. ladernarco. MD, MPH. Director: Ccnter?rr Epidmtiai'a? ?nd 11.1%me Sender: MMWR Editorial and Produetion Staff {Weakly} Charlotte K. Kent. WH, Acting Edits?): Cliff: Executive Editor Jacqueline Gindlei. MD. Edfmr Mary Dott, MD, MPH, Undue Editor F. Rutledge. Managing Miter Douglas W. Weatherwax. Lead Titania! Wm?Er?mr Glenn Damon. Soumya Dunworth, Term; Hood. MS. monitor Writer-damn Martha F. Boyd, Ema! Visual In?rmatian Specialist Maureen A. Lei-shy. Julia C. Martini-ore. SmphenR. Spriggs, Tong Yang. Wmdi Saarinen: 'Quang M. Dean. MBA. H. King. Terraye M. Starr, Mona Yang, In?rmerian Titans/95y Summits: MMWR Editorial Board Timothy F. Jones. MD. Charmer? Matthew L. Bonitoo. MD. MPH Virginia a. Gains, Mt) Katherine Daniel. P110 Jonathan E. Fielding, 1MB, MPH, NIBA David Fleming. MD hJ MMWR June-3.2018 r' Vol.6? I No.22 William E. Helper-in. MD. Dunn. MPH King K. Holmes. MD, Robin Ikrda. MD. MPH Rims F. Kitabhaz. MU Phyilis Meadows. rho, MSN, RN lewd Mullen. MD, MPH. MFA Jeii'Niedes-deppe. Patricia Quinlisk. MD. MPH Patrick 1.. Remington. MD. MPH Carlos Roig. MS. MA L. Roper, MU, son William Sellai?ler. MD US Department of Health and Human Servicesr'Cenlers for Disease Control and Prevention PROOF PROOF PROOF PROOF Morbidity and Mortality Weekly Report per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employ- ing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes. Characteristics of persons aged 1:10 years who died by suicide. with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 27 states? with complete data participating in CDC's National Violent Death Reporting System in 2015. defines mental health conditions as disorders and lisred in the DiagnosticandSratirrimlMammi aneruaz?Dirom?err, Ef?e Edition with the exception of problematic alcohol use and other substance use that are captured separately in VDRS aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxi- cology), and law enforcement reports. A range of circumstances (relationship problems, life stressors, and recent or impending crises) have been identified as potential risk factors for suicide in Circumstances captured are those identi?ed as cone tributing to suicide in coronerfmedieal examiner or law en force menr reports, which reflect information provided by family and friends at the time of death. Decedents could have experienced multiple circumstances. .Decedents with and without known mental health conditions were compared using chirsquare tests. Logistic regression analyses were used to estimate adjusted odds ratios (aORs) with 95% confidence intervals (Cls), controlling for sex, age group, and racelethniciry. Results The most recent overall suicide rates (representing 2014? 2016) varied fourfold, from 6.9 (District of Columbia) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from 0.8 per 100,000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases >30?fi] observed in 2?5 states (Supplementary Table; (Figure). Modeled suicide rate trends indicated significant increases in 44 states, among males (34 states) and females (4.3 states), as well as for the United States overall (Supplementary Table: Nationally, the model? estimated average annual percentage change for the overall Alaska. Arizona. Colorado, Connecticut. Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, Newjetsey, New Mexico. New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island. South 'Carolina. Utah, Vermont. Virginia, and 1 i'v'isconsin. US Department of Health and Human Services/Centers for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 4w I I ncreasez31%?3?% ncrease119%?30% Increase; 6%-18% 1:1 Decrease: TWO Per 100,000 population, age?adjusted to the 2000 1.1.5. standard population. suicide rate was an increase of1.5%. By sex, estimated national rate trends further indicated significant average annual per? centage change increases for males and females (SupplementatyTahle; Suicide decedents without known mental health conditions (1 1,039; 54.0%) were compared with those with known mental health conditions (9,402; 46.0%) for 27' states. Whereas dece- dents were predominantly male (?liable 1) and non- Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 63.8%; odds ratio 2.3, 95% CI 2.2?2.5) and belong to a racialr?ethnic minority (OR range Suicide decedents without known mental health conditions also had significantly higher odds ofperpetrating homicide followed by suicide (30R 2 2.9, 95% C1 Among decedents aged 3:18 years, 20.1% of those without known mental health conditions and 15.3% ofthose with mental health conditions had previously served in the U.S. military or were serving at the time of death. Whereas firearms were the most common method ofsuicide overall decedents without known mental health conditions were more likely to die by firearm and less likelyr to die by or poisoning than were those with known mental health conditions 31.3%, and 19.8%, respectively). These differences remained signi?cant in the adjusted models. Toxicology testing was less likely to he performed for dece? dents without known mental health conditions. Among those with toxicology results, decedents withoth known mental health MMWR r' June 8. 2018 Vol.6? r' No.22 3 PROOF PROOF PROOF Morbidity and Mortality Weekly Report TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National lii'iolent Death Reporting System. 22 states.* 2015 Known mental No known mental Total health condition1 health condition 20.440} in 9.40?) in 1 1.11391 ChiAsquare 0R5 Adjusted oa'i Characteristic 010.1%} 010.1%) {110.1%} p-value {95% Cl} [95% Sex Male 15.202 {26.8} 6.459 (68.8} 9.233 {83.6) {0.01 2.3 NA Female 4.244 {23.2} 2.938 (31.2} 1.306 {15.4} <0.01 0.4 NA Age has)? 10?24 2.304 {13.2} 1.211 {12.9} 1.593 {14.4) <0.01 1.1 {1.14.2} NA 25?44 15.456.81.15] 3.036 {32.3} 3.420 {31.0} ?:0.05 0.9 NA 45-64 1.21:! {31.7} 3.320 {40.6} 3.893 {35.3} (0.01 08103?03} NA 2155 3.463 {12.0} 1.340 {14.2} 2.128 (19.3] (0.01 1.4 NA HacefEthnicity White. non?Hispanic 12.102 {83.6} 8,165 {36.3} 8.93? {81.0} (0.01 0.5 NA Black. non-Hispanic 1,225 411 812 <0.01 1.2 NA American Indianmlaska Native. 328 112 266122.41 ?c001 2.0 NA non-Hispanic Asian. non~Hi5panic 526 235 341 <0.05 1.2 NA Hispanic 1.096631} 463 533 <0.05 1.2 NA Other 66 {03} 21 {0.21 45011.4] (0.05 1.3 NA Extended demographics Ever served in military? 3.429 {17.3} 1.3541153] 2.025 {20.1} <0.01 1.4 1.1 Homeless 240 104 {1.11 136 {1.31 NS 1.1 1.210.9?151 Single suicide 20.063 (98.2] 9.313 (99.1] 10.245 {92.4} (0.01 0.3 0.4 Homicide followed by suicide 319 64 {0.71 255 {2.31 <0.01 3.5 231122?33) Multiple suicides 64 25 39 NS 1.3 1.6 Method Firearm 9.909 {43.5} 3.821 {40.6} 6.088 {55.3} <0.01 1.8 1.6 5.907 {28.9} 2.940 {31.3} 2.96? {26.9} ?5.0.01 0.8 0.3 Poisoning 3.003 {14.7} 1.351 (19.3] 1.142 {10.4} (0.01 0.510.4?05} 0.6 Substance class causing cleathg'? Other otter?the?counter} 1.021 {34.0} 066 {35.8} 355 {31.1} {0.01 0.13 0.9 Upioids 944 {31.4} 608 {32.2) 335 {29.4) NS 0.9 0.9 Antidepressants 300126.61 044 {34.61 150 {13.21 (0.01 0.3 0.3 Benzodiazepines 624 {20.8} 458 {25.1} 156 {13.7} {0.01 0.5 {0.4n0.6} 0.5 219 195 {10.5} 24 40.01 0.2 {0.1413} 0.2 Other 1.5951113} 13018.31 815112.41 ?10.05 0.910.840] 09103?10} See table footnotes on next page. conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids laC?R 0.90. 95% CI but were more likely to test positive For alcohol 1.2. 95%. Ci Information on circumstances surrounding suicide were available For all decedents with mental health conditions and approximately 35% of those without known mental health conditions {9.557) in 2? scares (Table Persons without known mental health conditions were less likely to have any problematic substance use 0.7. 95% Cl 0.7?0.8) than were persons with lcnown mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment iust over half were in treatment at the time oi: death. Decedents without known mental health conditions had a significantly higher likelihood ofany relationship problem/loss st MMWR June8.2018 r' Vol.6? 1 No. 22 than did those with known mental health conditions speci?cally intimate partner problems {30.2% ver? sus argumentsicon?icts {17.5% versus and perpetrating interpersonal violence in the past month versus Decedents without known mental health condi- tions were also more likely than were those with known mental health conditions to have experienced anyliie stressors (50.5% versus 412%} such as recent criminal legal problems (10.7% versus or evictioniloss ol?home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks. respectively] crisis (a current or acute event thought to contribute to the suicide) (32.9% ver- sus 213.0%] . All ofthese diffierences remained signi?cant in the adjusted models. Physical health problems and iobr?financial problems were commonly contributing stressors among both persons without mental health conditions {23.2% and 15.6%, respectively} and those with mental health conditions {21.4% US Department of Heaith and Human Servicestenters for DiseaSE Control and Prevention PROCH3 Morbidity and Mortality Weekly Report PROOF PRUCH3 TABLE 1. {Continued} Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 22 states,* 2015 Known mental No known mental Total health condition'r health condition 20,445} {n 9,402} {n 11,039} Chi-square 005 Adjusted Characteristic No. No. No. {it} p-value {95% Cl} {95% Cl} Toxicology results Any toxicology testing 1 3.3120511} 6,653 {20.8} 6,659 {50.3} (0.01 0.5 {0.64121 Positive for at substance? 9,913 {114.4} 5,192 {18.0} 4,721 {20.9} :00] 0.7 0.8 Substance detected'? Alcohol Tested 10,950 {53.6} 5,409 {57.5} 5,541 {50.2} (0.01 0.7 0.3 Positive 4,442 {40.6} 2,115 {39.1} 2,327 {42.0} <0.01 1.1 1.2 Upioids Tested 0,554 {41.3} 4,258 {45.3} 4,296 {38.9} <0.01 0.0 {0.74113} 0.8 2,229 {20.6} 1,233 {29.1} 1,041 {24.2} {0.01 0.3 0.9 Benzodiazepines Tested 3,124 {39.2} 4,225 3,898 {35.3} {0.01 0.2 0.7 Positive 2,464 {30.3} {38.8} 325 {21.2} <0.01 0.4 0.5 Cocaine Tested 192311390} 3,866 {41.1} {0.01 0.9 0.9 Positive 499 216 203 <0.05 1.2 1.2 {1 .0-1 AmphetamirIEs Tested 1,615 {32.2} 3,095 {39.3} 3,919 {35.5} (0.01 0.9 0.9 {0.3419} Positive 236 326 {10.2} 350 NS 0.9 1.0 {0.8?1 Marijuana Tested 6,569 {32.1} 3,127 {33.2} 3.442 {31.2} (0.01 0.9 0.9 Positive 1,471 {22.4) 211.1{222} 261 {22.1} NS 10 {0.94.13 0.9 Antidepressants Tested 5,425 {26.5} 3, 03 {33,0} 2,3 22 {21.0} {0.01 0.5 {0.5416} 0.5 Positive 2,21 4010.3} 1,735 {55.9} 429 {20.6} <0.01 0.2 0.2 Abbreviations: Ci con?dence interval; NA not adjusted: N5 not signi?cant; OH odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New Vorit, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin, Decedent had been identi?ed as having a current diagnosis of mental health condition in coroneri'medical examiner or law enforcement reports. CiFt reflects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for sex, age group, and racefethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. ii Denominator is decedents aged 213 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominatoris decedents with any toxicology testing. Denominator for each positive group is the number tested for the substanca in that group. and 16.8%, respectively}. Similarly, among all persons with recent crises. intimate partner problems were the most com- mon types and did not difter by group. Decedents withour known mental health conditions had signi?cantly lower odds of recent release From any institu? tion 0.5, 95% Cl Among those recently released, without known mental health conditions were significantly more likely than decedents with mental health conditions to have been released from a correctional Facility (25.7% versus hospital {43.7% versus or other facility. such as an alcoholfsubsrance use treatment facility {24.2% versus Among decedents with known mental health conditions who Were recently released from an institution. 46.7% were released from faciliti?s. Decedents without known mental health conditions were sig- ni?cantly less likely to have a history of suicidal ideation US Department of Health and Human Servicesr?Centers for Disease Control and Prevention or prior suicide attempts compared with those with known mental health conditions {40.3% and 29.4%. respixtively}. Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health conditions, respectively. Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states. and 25 states experienced increases 330%. Rates increased signi?cantly among males in 34 states, and Females in 43 states. Additional research into the speci?c causes of these trends is needed. Data from the 27 states participating in provide important insight into cir- cumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not cauSed by a single factor however, suicide prevention is often MMWR June 8. .2018 Vol.6? r' No.22 f1 moor Psoor 00001: 13121001: Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged 1:10 years with and without known mental health conditions National Violent Death Reporting System, states,* 2015 Known No known mental health mental health Chir condition?l condition square 0115 Adjusted 0111 Characteristic Total No. No. p-value {95% {95% Cl} Suicide with known circumstances 18.264 {91.81 9,407 {1001 9.357 {34.81 <0.01 NIA Mental health Any current diagnosed mental health condition? 2.076 {2521 WA MA MA MA Anxiety disorder ?t1 1.519 {10.01 WA MA MA Bipolar disorder 1,431 {15.21 000 MA NM NM Schizophrenia 509 {5.41 MIA MA MA Ptso ?tt 424 {4.51 MA MA not 1* 226 {2.41 MIA WA MA Not specified 1'60 {8.11 are Current depressed mood?? 7.0331315} 3,952 {42.11 300013191 e001 {00-00} arias-0.11 Problematic substance use Any 5.3191233} 2,926 {31.61 2.343 {25.01 <0.01 0.2 0.2 Alcohol 3.2681114} 1,862 {19.81 1.406 {15.0} 4:001 07103-081 Other 1,316 {14.1} <0.01 {0.7?0.01 0.7" Treatment Current mental healthr'substance use treatment 5.141 {22.41 5,077 {54.01 64 <0.01 0.01 {0.01~0.011 0.01 {0.01?0.01} Eyer treated for mental healt hrsuhstance disorder 6.71 {35.8} 6,323 {62.2} 394 {0.01 0.02 {0.02?0.02} 0.02 {0.02?0.03} Relationship problemsrloss Any relationship problemr'loss 194014241 122689.61 4,222 {45.1} 3.1 1.3 Intimate partner problem 5.0901212} 2,220 {24.11 2,323 {30.2} {0.01 4{1 3?1.5} 1.4 Perpetrator ofinterpersonal violence in past month 41412.2} 131 {1.41 283 2 .1'1 2.011.624} lliictim of Interpersonal violence in past month 3410.41 5301.61 31 {0.05 6{0. 4?0. 91 0.0 {0.5?1 Family relationship problem 1.621 {3.91 1323 {9.31 295 NS 1. 0} 1.0 {0.9?1 .11 Other relationship problem lnonintimatel 40312.1} 202 {2.11 201 {2.11 NS 1.0 1 21 1.1 Argument orconflict {not specified) 2.9141155} 1.278 {13.61 1.636 {17.51 <0.01 1.3 Death of a loved one {any} 1,497 [3.01 326 {13.31 621 <0.01 0.3 0.9 {0.81?0.91 Nonsuicide death 1,181 [6.31 641' {6.91 534 (0.01 0.3 0910.34.01 Suicide of family or friend 32912.0} 212' 162 (0.01 {0.6e0_9} 08 (0.1L 1 .01 Other life stressors Any life stressor 9.171 {43.91 4,442 {42.21 4,729 {50.51 6.0.01 1.1 1.1 Recent criminal legal problem 1.588185} 586 1,002 {10.7} (0.01 1.8 Other legal problem 740 {4.01 323 {4.01 320 {4.01 NS 1.0 {0.8?1 1.0 Physical health problem 4.1291223} 2.012(21011 2.167 {23.2} 4:10.01 1 1.0 Jebrnnanciel problem? 2,941 {15.21 1,530 (10.31 1.411 {15.01 <0.05 0.9 0.9 Eviction orloss ofhome 72213.01 31213.4] 405 <0.01 1.3 1.4{1.2v1.61 School 162119.91 70 {12.131 512 {21.9} NS 9? 1 .81 1.3 {0.9?1 Recent release from an institutrentn? 1.412 941 {10.21 471 <0.0r 5.10 4?0 51 0.5 laillPrisonr?Detention facility 203 {14.41 32 {8.21 121 {25.2} {0.016.{2 2?4. 91 4.5 Hospital 512136.61 311 {33.0} 206013.21 <0.016.{1 3? 2 0} 1.3 hospitalr'institution 469133.21 439016.71 3016.41 (0.01 0.1 {0.1 11 0.1 {0.1?0.11 Other {includes alcoholr?SU treatment facilities} 223115.31 109111.51 114 {24.2} <0.01 2.4 {1 .8-33} 2.511.803] See table footnotes on next page. oriented toward mental health conditions alone with regard to identi?cation of suicidal persons, treatment of mental health conditions, and prevention of Th is study Found that approximately haltofsuicide in did not have a known mental health condition. indicating that additional focus on health Factors Further upstream could provide important information for a public hcold't approach (It?ll). Those without a known mental health condition suffered more from relationship problems and other life such as criminalflegal matters, cvictiom?loss othomc. and recent or impending criscs. 6 MMWR a? June 8.2018 I No.22 Similarly. persons with mental health conditions also often experienced other circumstances such as relationship problems and jobf?nancial or physical health problems that contriburcd to their suicide. These ?ndings point to the need to both prevent the circumstances associated with the onset of? mental health conditions and support persons with known mental health conditions to decrease their risk for poor outcomes Two d'iirds of suicide with mental health conditions had a history of treatment For mental health or substance use disorders, with approximately halFin treatment when they died. This ?nd? ing suggests need For additional safety supports. including US Department of Health and Human Servicesr?Centers for Diseaso Controi and Prevention PROOF PROOF Morbidity and Mortality Weekly Report HOG TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 27 states,? 2015 Known No known mental health mental health Chi- condition+ condition square 0R5 Adjusted 0R1 Characteristic Total No. No. p?value (95% Cl} [95% Cl} Crisis within past or upcoming 2 5,525 {29.4} 2,444 {26.0} 3,081 {32.9} c??l 1 4 Intimate partner problem 1,968 {35.6} 854' {34.9} 1,1 14 {36.2} NS 1 1 1 .1 {119?1 .2) Physical health problem 139113.41 3151123} 424 (13.3} NS 1 1 1-0 Criminal legal problem 521 [11.2) 203 4181:1311} (0.01 1.6 {13-131 Family relationship problem 430 [18} 212 218L111 ?(0.05 8 0.9 [03?1 Job problem 354 191 153 corn 7' 0.7 Suicide evantfhistory Left a note 6,463 [34.51 3,132 313685.11 N5 1.1 {1.0?1 1.2 Disclosed suicide intent 4,4115 [23.5) 2,306 [24.5] 2.099 {22.4) (GET 0.9 {0.8?1 0.9 History of ideation 5,990 (31.9} 3,838 (413.8] 2,152 cili'l'r I14 0.4 (0.4-0.5) History of attempts 3,732 [19.91 2,770 {29.41 962 [10.3] ?lm [1.3 0.3 {1.13?0.31 Abbreviations: ADDEADHD attention de?cit disorderrattention deficit hyperactivity disorder; Cl 2 con?dence interval; MIA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; St] substance use. Alaska. Arizona. Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan. Minnesota, New Hampshire. New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. 0R re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for sex. age group. and racerethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclUsive. Therefore, sums of percentages for the diagnoSed conditions exceed 100%. Denominator includes the number of deCEdents with one or more current diagnosed mental health conditions. The specific type of mental health condition was calculated only among those with one or more known diagnosed mental health conditions. 55 Not a diagnosis. Denominator is decedents aged 218 years. Denominator is decedents aged 10?1 3 years. 111 Denominator ofinstitution subgroup is decedents with recent release from an institution- Recent release from an institution is de?ned as having occurred within the past month. 5% Denominator of crisis subgroUp is decedents with any crisis within past or Upcoming 2 Weeks. Crises depicted here represent the most commonly occurring categories, broader implementation of affordable and effective treatment modalities, such as doctor?patient collaborative car: models and proven cognitive-behavioral therapies. In addition, increased access to behavioral health providers in areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prcven~ tion and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range effectors contributing to suicide. Prevention strategies include strengthening economic supports (cg, housing stabilization policies. household ?nancial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially earl}:r in life; promoting social to increase a Sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans. persons with physical! mental health conditions) Other strategies incl udc creating protective environments (cg, reducing access to lethal means among persons at risk for suicide, creating organizational and workplace policies to promote help-seeking, easing transitions into and our ofworlc for persons with mental health conditions US Department of Health and Human Services/Centers for Disease Control and Prevention and other life challenges), strengthening access to and delivery of care, supporting family and friends after a suicide, and encourag- ing the media to follow Safe reporting recommendations Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention (30). The ?ndings in this report are subject to at least three limita? tions. First, in the starerlevel analysis, rankings for four states (Maryland. Massachusetts, Rhodc island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time {porentially biasing estimated rare trends upward}. Second, is not yet nationally representative; the 27 States included represent 49.6% ofthe population census .govf pages.f ll . Finally, abstractors of data are limited to informa? tion contained in investigative reports. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies that include more interviews with nextrof?lcin often identify greater attributions to mental health disorders however. many methodological variations across studies exist it is likely that some persons without known mental 'Mi MMWH r' June 8. 2018 Vol.6? r' No.22 PROOF PROOF ROOF PROOF Morbidity and Mortality Weekly Report Summary What is already known about this tapic? In an 6. nearly 45.000persons died by suicidein the United States. Mental health conditions are one ofseveraleontributurs-to. suicide. What is added by this renort? During 1999?2615. suicide rates increased in nearly every state.- including )3an increases irr25 data from 27 states indicate 54% of suicide decedent's Were not known to have Mental health conditions. Relationship. substance use. heaith. and job or ?nancial problems are among the other circumstances contributing to suicide. What are the no pllcations for public health practice? A-comprehemive approach using proven prevention strategies. such asthosein Technical Package for Suicide Prevention, can help readt the national goal of reducing the annual suicide rate 20% by 2025. health conditions in the current study were experiencing mental health challenges that were unknown?or not reported by key informants. Nonetheless, the high prevalence of diverse con? tributing circumstances among those with and without known mental health conditions suggests the importance ofaddressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk facrors are avail- able. States and communities can use data from and resources such as Preventing Suicide: A Fromm! Portage (finality. Programs. and Practices to better under- stand suicide in their populations, prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgments Robert Anderson. Holly Margaret Warner. Division of Vital Statistics, National Center for Health Statistics. CDC. Con?ict of Interest No conflicts of interest were reported. 'Division of Prevention. National Center for Injury Prevention and Control. CDC: zDivis?ion of Analysis. Research. and Practice Integral-ion. National Center for [niury Prevention and Control. CDC. Corresponding author: Deborah M. Stone. dstonrd?-cdcgov. 73304886942. El MMWFI i Vol.6? i No.22 lull 6. 9. ID. ll. I4. References .CDC. Web?based lniury Statistics Query and Reporting System Atlanta. CA: National Center for Injury Prevention and Control: ZUIB. . hey?Stephenson AZ. Crosby AE. Jack SPD. T. Kresnow- Seder-ca Suicide trends among and within urbanization levels by sex. raccicthnicity, age group. and mechanism ofdcath?Uniled States. 2001?2Ul5. MMWR Surveili Summ 201?;66lNo. SS- orgfi D. 5585imn1wrss??l Sal . Curtin SC. Warner M. l?l. Increase in suicide in the United States. 1999?2l?r14. NCHS data brief no 241. Hyarrsvillc. MD: US Department of Health and Human Services. CDC, National Center for Health Statistics; 2016. db24l.pc f . Kochaneit K. Murphy 5. Xu J. Arias E. Mortality in the United States. 20] NCHS data brief no 2.93. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 201?. . Office of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action. \Vashington. DC: US Department of Health and Human Services. Office of the Surgeon General: 2012. surgeongcneralgovil ibraryi rioni Zalsrnan G. Hawton K. D. er Suicide prevention strategies revisited: Iii-year systematic review. Lancet 20 D. [0 1 61"5221 5?0366i .Torguson K, O?Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; ZUIT. .Crcpeau?Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 1 American Association. Diagnostic and statisrical manual of mental disorders Washington. DC: American Association; 20H. Cainc ED. Reed]. Hindrnan J. Quinlan K. Comprehensive. integrated approaches to suicide prevention: pracrical guidance. Inj Prev 2017. Epub December 20. 2131?. .World Health Organization. Risks to mentai health: an overview of vulnerabilities and risk factors. Geneva. Switzerland: World Health Organization: 2012. Stone DM. Holland KM. Bartholow BN. Crosby AE. Davis SP. Wilkins N. Preventing suicide: 3 technical package ofpolicy. programs. and practice. Atlanta. GA: US Department of Health and Human Services. CDC: 2017. . Milocr A. Svcticic Dc- Lco D. Suicide in Lhe absence disorder? A rcvicvi.r of autopsy studies across countries. Int] Soc i 0.1 Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 0. US Department of Heaith and Human ServicesiCenters for Disease Control and Prevention Rev 3.26.18, {learance- v97; Short title: Circumstances to Suicide and increasing Trends in State Suicide Rates ameng?peeple?a?LQ?years Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, PhD;l Scott R. Kegler, Kerning Yuan, Kristin M. Holland, Asha Z. lvey-Stephenson, Alex E. Crosby, M01 Structured abstract [2451'250 words?this word count is not included in the 1800 max for the remainder) Background: Suicide rates have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three~year periods {19992016}, were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C (012.). Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known MHP. Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of 30% or more. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, those without known MHP were more likely (all p303.? than those with a MHP to have relationship problems?oss (45.1% vs life stressorsg?ioss (54.2% vs 49.7%} and recentlim pending crises (32.9% vs but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE (255,050 words) in 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States (U.S.), among people 1310 years old Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self?harm injuries cost the nation more than $69 billion in direct medical and work loss costs The Notional Strotegyfor Suicide Prevention(N55P) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems (MHP) Other contributing circumstances include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problem? solving skills, and prior suicide attempts [51. Expanded awareness of the additional circumstances that Rev 3.26.18, e-gIeaI?aI-Ice-va?s?. contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?Ievel comprehensive suicide prevention. words} Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in younger children are variable Age?specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationoi Ciossificotion of Diseases 10!? Revision, underlying-cause-of death codes X84, Y87.0, Agemspecific population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three?year aggregate periods spanning 1999?2016. Rate estimates were age?adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics (Table 2} and circumstances (Table 3} of suicide decedents 210 years, with and without known MHP, were compared in the 2? states with complete data participating in National Violent Death Reporting System in 2015. de?nes MHP as disorders and svagudromgslisted in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition With the exception of alcohol and other Substance dependence, vilnch are caotured separately in . Tier?anissteeiy, alcohol?and aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology}, and law enforcement reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and race/ethnicity. RESULTS (Saga/600 words} The most recent overall suicide rates (representing 201442016} varied four?fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (with the 9th highest current suicide rate}, with absolute increases ranging from +0.8 {Delaware} to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +5.91% {Delaware} to {North Dakota}, with increases more than 30% observed in 25 1 Figure 11. Modeled suicide rate trends indicated significant increases for 44 states, as? velias for the s- over-ail (Table i} and for males (34 states} and females {43 states} as well as for the eye-fall (Table ll?By?sex?enedeieel?Fase memamm. Nationally, the model estimated AAPC for the overall suicide rate was By sex, the national AAPC was +11% for males and +26% for females. Suicide decedents without known MHP (N211, 039} were compared to those with MHP (N29, 407,- ':'Ill col-sou are and IogIstIc repressIcn analyses sIgnificarIt. ?t I.p< 05h: _While all decedents were predominately male (Table 2; 76.8%} and non?Hispanic white those without known MHP. relative to those with MHP, were more likely male (83.5% vs. 63.8%; adjusted odds ratio 95% Cl 2.2-2.5} and racial/ethnic minorities (odds ratio range: 1.2-2.1; 95% Cl range Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide 2.9, 95% Cl of firearm suicide 2 111 112 113 114 115 116 117 118 119 I20 121 122 123 124 125 Rev 3.26.18, e-glearance-v?eE 1.6, 95% CI and of testing positive for alcohol 1.2, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. Although firearms were used most often; overall decedents with known MHP were more likely to die by suffocation (31.3 vs. 26.9%} and poisoning (19.8% vs. 10.4%} than those without known All suicide decedents with known MHP and approximately 85% without MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP, versus those with known MHP had a significantly greater likelihood (9:.pr any relationship problemj'loss (45.1% vs. specifically intimate partner problems (30.2% vs. arguments/conflicts (17.5% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other life stressors (54.2% vs such as criminal? legal problems vs. or evictionlloss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. Among both groups, the most common crises were intimate partner (36.2% vs. 34.9%} and physical health problems (13.8% vs. respectively. Decedents without known MHP had significantly lower odds of recent release (30920.5 952's CI from any institution, but among those who were recently released were significantly more likely to be released from a correctional facility (25.2% vs. or hospital (43.7% vs. 33.0%} than those with a known MHP. Among decedents with known MHP who were recently released from an institution gwqro?ugwere released from facilities. Decedents without known MHP, compared to those with MH P, were less likely to have a history of suicidal ideation 95% and prior suicide attempt 95% Bit/lore than 1 in five people in both groups disclosed suicide intentfreeuentlye (22.4% vs. Conclusions and Comments (619301700 words} From 1999?2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. research into the causes of these trends is necessary flirt. Fortunately, data from the stags in NW 8.5: can. shed on. the seaweed Hie?9%. 99% flaweecircum?ta nces That. so and help actiyities, staking; Researchers and practitioners regularly state that suicide is not ca used by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MH P, two-thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep 3 126 127r l28 129 130 131 132 133 134 135 136 137 138 139 140 141 142 i43 144 145 145 147 14s 149 150 .151 152 153 154 155 ?156 157 158 159 160 161 162 163 164 165 166 167 Rev 3.26.18, edglearance vie-2 them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, relationship, and/or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote he p~seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent decreasing suicide rates}, or decreased percentages of such deaths over time increasing suicide rates}. Second, is not yet nationally representative. the 27 states included in the current study represent 49.6% of the U.S. population. Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to MHP-erseseesaswee?aeew seearaeerernentm disorders. however many methodological variations across studies exist it is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence unreported by key informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of addressing the Lg'cgtirange of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gg 770-488-3912 168 169 120 171 172 173 174 175 176 177 178 179 .1813 181 182 183 184 185 .186 187 188 189 190 191 192 193 194 195 I96 19?:r 198 199 200 201 202 203 204 205 206 207 208 1209 Rev 3.26.18, {Jeerance- Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, i?Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. 10. 11. 12. 13. 14. 15. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. Nev-Stephenson, A.Z., et 31., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017'. 66118}: p. 1-16. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, MD. Kochanek, K., et al., Mortaiity in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: gaais and objectives for action: a report of the US Surgeon Genera:1 and of the Nationai Action for Suicide Prevention. 2012. Rosenman, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 159(2): p. 100102. Torguson, K. and A. O?Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 2017: Washington, D.C. Crepeau?Hobson, F., The Autopsy and Determination ofChiid Suicides: A Survey of Medicai Examiners. Archives ofSuicide Research, 2010. 1411}: p. 24?34. American Association, Diagnostic and statisticai manuai ofrnentai disorders 2013: American Pub. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life? Threatening Behavior, 2002. 32(51): 4959. Stone, D.M., et al., Preventing suicide: A technicai package of poiicies, programs, and practice. 2017. World Health Organization, Risks to mentai heaith: An overview of and risk factors. Geneva: WHO, 2012. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. injury prevention, 2017: p. injuryprev-2017?042366. Milner, A., J. Sveticic, and D. De Leo, Suicide in the absence ofrnentai disorder? A review autopsy studies across countries. [nt Soc 2013. 59(6): p. 545?54. Pouliot, L. and D. De Leo, Criticai issues in autopsy studies. Suicide Life Threat Behav, 2006. 36(5): p. 491-510. Attachment: Stone Suicide Vital Signs 3.26.155. {intuit-s. Fla. Word Count: 1713/1800 Lth-J 45- I'Rev 3.26.18, e-clearance Short Title: Vital Signs: ontributing Circumstance-s to Suicide and Increasing Trends in State Suicide Rat - Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Structured abstract [2115,1250 words?this word count is not included in the 1800 max for the remainder} Background: Suicide ratesEn the United StateihaVE risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state?level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged :10 years, by state and sex, across six consecutive three-year periods (1999?2016}, were assessed using data from the National Vital Statistics System for 50 states and Washington, DC Data from the National Violent Death Reporting System, covering states in 2015, were used to examine contributing cirCumstances among decedents with and without known Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of P096 or half of decedents did not have a known MHP. Among decedents with circumstance information, those without known MHP were more likely {all p501} than those with a to have relationship problemsi'loss {45.1% vs life stressors (54.2% vs 493%] and recentfimpending crises {32.9% vs but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [ageradjusted? occurred in the United States among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racial/ethnic grows, and urbanization levels Suicide is the 10?1 leading cause of death and is one ofjust three leading causes that are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm,a key risk factor for suicide, increased nearly 415% betWeen 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strategy for Suicide PreventioanSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societall. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing circumstances include social and economic problems, access to lethal means leg, substances, firearms, bridges] among people at risk, poor coping and problem- solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of Comment [so k6]: This title seems in opposition to the order of our presentation framework first we present *what" - the trends - and then we talk about *why* - the contributing ,circumstances. Comment Without this, the contextfgeography isn't fully set. The MMWR editors might want and the study years included in the title at ,some point. Comment Isnk?]: This important "factoid" appears in three places in this report. The phrase ?30% or more" Is technically correct, although the phrase ?more than 30%" is even more correct {the lower limit to the range for the 3'5 highest states is about 32% -- and no state is right at 30% even when rounding}. Regardless of which phrase we adopt, it seems that we should be consistent throughout. I made a suggested revision further below {in the Results} assuming that We are going with or more". ?i In. Rev 3.25.18, e-clearance 40 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific 4 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level 42 comprehensive suicide prevention. 43 METHODS 44 Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in yoUnger children 45 are variable Age?specific suicide counts were tabulated based on National Vital Statistics System coded death 46 certi?cate records (international Ciassrficotion of Diseases 10?? Revision, underlying-cause?of death codes X60- 47 X34, YBIG, 003}. Age-specific population estimates were obtained from US. Census Center for 48 Health Statistics bridged~race population data releases. 49 50 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 5] spanning 1999-2016- Rate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed 52 per 100,000 persons per year. Agewadjusted suicide rate trends were modeled using the same three-year data 53 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 54 are reported in terms of average annual percentage changes 55 56 Characteristics [Table 2) and circumstances (Table 3) of suicide decedents 210 years, with and without known 57 were compared in the 2? states with complete data participating in National Violent Death 58 Reporting System in 2015. defines as disorders and listed in the Diagnostic and 59 Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance 60 dependence, which are captured separately in aggregates data from three primary data sources: 6 death certificates, coronerfmedicai examiner reports {including toxicology), and law enforcement 62 reports. Decedents with and without known lv'tHP were compared using Chi?square tests. Logistic regression 63 analyses estimated adjusted odds ratios with 55% confidence intervals controlling for age group, sex, and 154 raceiethnicity. 65 RESULTS so The most recent overall suicide rates [representing 2014?2015] varied four?fold, from 6.9 to 29.2 at (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except bit Nevada (with the highest current suicide rate), with absolute increases ranging from +0.3 (Delaware) to +3.1 69 (Wyoming) per 100,000. Percentage increases in rates ranged from +59% (Delaware) to (North Dakota), I ?0 with increases 'il i2 Modeled suicide rate trends indicated significant increases for 44 states, for males (34 states} and females (43 73 states], as well as for the 0.5. overall (Table Nationally, the model-estimated AAPC forthe overall suicide 74 rate was By sex, the national AAPC was +11% for males and for females. ?5 76 Suicide decedents without known were compared to those with MHP all chivsouare ii? and logistic regression analyses significant at p505). While all decedents were predominately male (Table 2: and non-Hispanic white those without known MHP, relative to those with MHP, were more 79 likely male (83.5% vs. adjusted odds ratio 95% CI 2.2?2.5} and racialfethnic minorities (odds 30 ratio range: CI range Suicide decedents without known lv'lHP also had til significantly greater odds of perpetrating homicide-suicide 19,9596 CI of firearm suicide lac-R E42 33 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the 34 0.5. military. 30% or in 25 states [Table_1.__F_i_gure 1.5, 55% CI and of testing positive for alcohol 1.2, 95% CI Among adult decedentsComment For consistency. See comments above. 100 101 102 103 104 105 106 107 108 109 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 Rev 3.26.18, e-clearance Although firearms were used most often overall decedents with known MHP were more likely to die by suffocation (31.3 vs. 26.9%) and poisoning (19.8% vs. 10.4%} than those without known MH P. All suicide decedents with known MHP and approximately 85% without MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems 95% (2120.10.81. While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MH P, versus those with known MHP had a significantly greater likelihood of any relationship problemfloss (45.1% vs. specifically intimate partner problems (30.2% vs. (17.5% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other life stressors (54.2% vs such as criminal- legal problems (10.7% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. Among both groups, the most common crises were intimate partner (36.2% vs. 34.9%] and physical health problems (13.8% vs. respectively. Decedents without known MHP had significantly lower odds of recent release 95% Cl (040.5) from any institution, but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MH P, compared to those with MH P, were less likely to have a history of suicidal ideation 95% (2120405) and prior suicide attempt 95% More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Additional research into the specific causes of these trends is necessary. Fortunately, data from the states participating in can shed light on the circumstances that contributed to recent suicides and help guide prevention activities. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MH P, two?thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as job/?financial, relationship, andfor physical health problems. These findings point to the need to both prevent the conditions 3 Fiev 3.25.18, e-clearance 26 associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes I28 These results, together, underscore the importance of comprehensive statewide suicide prevention activities 29 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I30 economic supports le.g., housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especiallv early in 132 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, 133 emotional, and social Support, and identifying and better Supporting people at risk. Other strategies include I34 creating protective environments leg, reducing access to lethal means among people at risk, creating 35 organizational and workplace policies to promote help-seeking, easing transitions into and out of Work for I36 people with MHP and other life challenges}, supporting family and friends after a suicide, and assuring safe 13? reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning Iii-8 and implementing such a comprehensive approach to suicide prevention I39 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, I4ti might have been impacted by large proportions of injury deaths of undetermined suieide?r-ate-s ootentiallv biasing reported suicide rates downward}, or decreased percentages of such deaths I43 vet nationally representative; the 27 states included in the current study represent 49.6% of the US. population I44 Third, abstractors of data are limited to information contained in investigative reports. Therefore. I45 the extent of informant knowledge can impact data completeness and accuracy. Studies including more inudepth 146 interviews with next-of-kin often see greater attributions to mental disorders however many l4? methodological variations across studies exist It is likely that some people without known MHP in the I48 current studi,r were experiencing mental health challenges that were unknown, and hence unreported by key I49 informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of [50 addressing the broad range of contributing circumstances. 15] Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are I52 available. States and communities can use data from and resources such as CDC's Preventing Suicide: a 53 Technical Package of Policies, Programs, and Practices [11] to better understand their suicide problem and I54 prioritize evidence-based comprehensive suicide prevention. I55 Acknowledgments I56 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. I58 Con?ict of Interest No conflicts of interest Were reported. I50 lol Corresponding author: Deborah M. Stone, dstone3@cdc.gov amass-3942 I62 163 Author Affiliations: I64 1Division of Violence Prevention, National Center for Injuri,r Prevention and Control, 2Division of Analysis, 165 Research, and Practice Integration, National Center for Injury Prevention and Control, CDC I as References: I42 over time biasing estimated rate trends Comment Isnk?]: These alternate parenthetical phrases might be considered again. The current ones confused me when I ?rst read them, and although I Was able to determine what they were intended to convey, that might only be due to my participation in the discossion with NCHS. In the first parenthetical phrase, which basically refers to Maryland, it sounds like we?re saying that decreasing rates might be the anticipated result. But Marylands reported rates did n?t decrease they increased despite the large and persistent percentage of injury deaths with undetermined intent. They might not have increased as much as they would have with more complete reporting of intent, and if that is true then we are looking at an increase that is biased downward. Dr their reported rates might be biased downward across the board. For the second parenthetical phrase, which basically refers to Massachusetts, Rhode Island. and Utah, it sounds as though We might be attributing the increase In reported rates to the decrease in the percentage of iniury deaths of undetermined intent. But the observed increases, especially in Utah, might be too large to be explained by this alone. More likely, it Would have biased already Lincreasing rate trends further upwards. 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 Rev 3.26.18, e-clearance 1. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. 2. Ivev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. 3. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hvattsville, MD. 4. Kochanek, K., et al., Martaiity in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Genera.f and of the Nationai Action for Suicide Prevention. 2012. 6. Rosenman, Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 169(2): 100-102. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 2017: Washington, DC. 8. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicoi Examiners. Archives ofSuicide Research, 2010. 14(1): p. 2434. 9. American Association, Diagnostic and statisticai manuai ofmentai disorders 2013: American Pub. 10. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life Threatening Behavior, 2002. 32(51}: 49?59. 11. Stone, D.M., et al., Preventing suicide: A technicai package of poiicies, programs, and practice. 2017. 12. World Health Organization, Risks to mentai heaith: An overview of and risk factors. Geneva: WHO, 2012. 13. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicoi guidance. Injurv prevention, 2017: p. injurvprev?2017-042366. 14. Milner, A, J. Sveticic, and D. De Leo, Suicide in the absence of mentai disorder? A review of autopsy studies across countries. IntJ Soc 2013. 59(6): p. 545-54. 15. Pouliot, L. and D. De Leo, Criticai issues in autopsy studies. Suicide Life Threat Behav, 2006. 36(5): p. 491610. Attachment: Suicide Vital Signs 3.26.18 (Tables_Fig, e-clearance) Word Count: 177171800 Ftev 3.2918, e-ciearance Comment ltgs?i: I think we need a transition like this to avoid implying that the circumstances listed are the only ,significant differences. Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates 2 Deborah M. Stone, Thomas R. Simon Ph0;? Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 3 M551 Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Comment [tgsii]: You had changed this to .05. I don't think we need this in the abstract since we say they were signi?cant. Just saying this is consistent ,with the too. ?vk 4 Structured abstract [2451 250 words?this word count is not included in the 1800 max for the remainder] 5 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state?level trends in suicide and other contributing 7 circumstances can inform comprehensive state suicide prevention planning. 8 Methods: Trends in age?adjusted suicide rates among people aged :10 years, by state and sex, across six 9 consecutive three-year periods [1999?2016], were assessed using data from the National Vital Statistics System Ill for 50 states and Washington, DJ: Data from the National Violent Death Reporting System, covering 27vr states in 2015, were used to examine contributing cirCumstances among decedents with and without known [2 MHP. l3 Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases l4 of?SGS?-eemarew. Rates increased significantly among males and females, in 34 and 43 states, [5 respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance if) information, in;lpcingtnese-wilhoet Leonie-MHPwere-mom ??49er {all install sham tin-me 17 with a- IVleIF-to have-relationship vs 49.7%) and IS recent/impending crises (32.9% vs were amt: sully r?iore pinu'ip inc-st;- without 1 I9 but theseeusmre?teet-es-were common across groups. '12: 20 Conclusions: Suicide rates increased significantly across most states from 1999-2016. 1Various circumstances 2 contributed to suicides among people with and without known 22 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 23 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 24- after a suicide occurs. 35 INTRODUCTION 26 BACKGROUND AND PURPOSE 2? in 2016, nearly 45,000 suicides [15.6f100,000 [age?adjustedll occurred in the United States among people 28 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 29 urbanization levels Suicide is the 10th leading cause of death and is one ofjust three leading causes that 30 are increasing Additionally, rates of Emergency Department visits for nonfatal self-harma key risk factor 3 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 32 nation more than $69 billion in direct medical and work loss costs 33 The National Strategy for Suicide PreventioanSSP} calls for a public health approach to suicide prevention 34 with efforts spanning across multiple levels lie, individual, family/relationship, community, and societal]. Such 35 an approach underscores that suicide is rarely caused by any single factor, but rather, is mold-determined. 36 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing circumstances include social and economic problems, 38 access to lethal means substances, ?rearms, bridges] among people at risk, poor coping and problem- 39 solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that Rev 3.26.18, e-clearance 40 contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of Comment ??591: We can?t include this 4 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state?specific here because not all of the comparisons i were signi?cant. I added some text below 42 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi-level to address this. Comment [tgsS]: I realized that we split up the results. about methods and that our toxicology results are incomplete. suggest moving these down. 43 comprehensive suicide prevention. 44 METHODS 45 Suicide rates were analyzed for people aged :10 years only, as attributions of suicidal intent in younger children as are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death 4? certificate records {international Ciassr?'cotion of Diseases 10?? Revision, underlying-cause-of death codes K60- 48 K84, YEIO, Age-specific population estimates were obtained from US. Census BureaulNational Center for 49 Health Statistics bridged-race population data releases. 50 5 National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods 52 spanning 1999-2016. Rate estimates were age-adjusted to the U.5. year 2000 standard population and expressed 53 per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?year data 54 aggregates, employing weighted least squares regression with inverse?variance weighting. Modeled rate trends i i i 55 are reported in terms of average annual percentage changes 56 i 5? Characteristics [Table 2} and circumstances {Table 3} of suicide decedents 210 years, with and without known 58 MHP, were compared in the 2? states with complete data participating in National Violent Death 5' 59 Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and i 60 Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance 6 dependence, which are captured separately in MUD-RS aggregates data from three primary data sources: 62 death Certificates, coronerfmedical examiner reports {including toxicology}, and law enforcement 63 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression 64 analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and i 65 racer?ethnicity. as RESULTS 6? The most recent overall suicide rates {representing 2014-2010} varied four-fold, from 5.9 to 29.2 5 08 {Montana} per 100,000 persons per year {Table Across the study period, rates increased in all states, except i i 69 Nevada {with the highest current suicide rate}, with absolute increases ranging from +0.3 {Delaware} to +3.1 70 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% {Delaware} to 67.5% {North Dakota}, TI with increases timore than 30% observed in 25 states [Table 1, Figure T2 I 7?3 Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females [43 T4 states}, as well as for the 0.5. overall {Table Nationally, the model?estimated MPG for the overall suicide 75 rate was By sex, the national AAPC was +1 l-Ia loI malesi 1,151} ??d?'?ilggi IsFemales. i {i I -ll T8 Suicide decedents without known MHP were compared to those with MHP sleulilmm aI While all decedents were predominately male {Table 2; T9 80 26.8%} and non?Hi5panic white those without known MHP, relative to those with MHP, were more 81 likely male {33.6% vs. 53.8%; seemed-odds ratio 95% CI 2.2?2.5} and racialfethnic minorities {sees i i 82 33 significantly greater odds of rsne 0Ri range: 1.2?2.1; 95% CI range Suicide decedents without known lv'lHP also had erpetrating homicide-suicide {agi?ted odds ratio .IOR 2 2.9, 95% CI nF . I?I?ii? 34 - .. 100 101 1102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 .12] 122 123 124 Rev 3.26.18, e-clearance Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. Weeghrv?ilgfirearms were the m_o_st common method overall both groups, decedents without known were more likely to die by firearm {55.3% vs1 40.65131 and less likely to die bysuffocation grand-poisoning (E13153 than those without known MHP. These differences remained significant in the adjusted models. Decedents Without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall laOR:0.8, 95% 0:07?08] but more likely to test positive for alcohol laOR21.2, 95% All suicide decedents with known MHP and approximately 85% without known had available circumstances information {Table 3). People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP, versus those with known MHP had a stgatficaatly?greater likelihood l?e-eng?l?of any relationship problem/loss (45.1% vs. specifically intimate partner problems (30.2% vs. arguments/conflicts [17.5% vs. and recently perpetrating interpersonal violence vs. They were also sig-rii?sao-tly more likely to have experienced other life stressors (54.2% vs such as criminal? legal problems (10.7% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of significant in the adjusted models. Among thosewith crises; intimate partner and physical health problems were the most common types for both groups and did not differ between them. Anreegbotngeeuosrthe-mest eon'lmoo arises were inhmate partner vs- 34 arid physicaJ- health problems {13.6% vs. 12 Decedents without known MHP had significantly lower odds of recent release (a0R=0.5 95% CI from any institution, but among those who were recently released they were significantly more likely to be released from a correctional facility {25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MH P, were less likely to have a history of suicidal ideation 95% and prior suicide attempt 95% More than 1 in five people in both groups disclosed suicide intent {22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 39%more than 30%er?aaere. Additional research into the specific causes of these trends is necessary. Fortunately, data from the states participating in can shed light on the circumstances that contributed to recent suicides and help guide prevention activities. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more 3 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 161 162 163 164 165 166 Rev 3.26.13, e-clearance relationship problems and life stressors such as criminal-legal matters, eviction/?loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MH P, two-thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, relationship, and/or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to Prevent suicide contagion Some states, such as Colorado. are planning and implementing such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of' Injury deaths of undetermined intent-Haead-ec-reasieg suierde?Fai?esl, potentially biasing reported suicide rates downward] or decreased percentages of such deaths over time (Le. biasing estimated rate trends upward]. Second, Is not yet nationally representative; the 27 states included in the current study represent 49.6% ofthe U.S. population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in?depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence unreported by key informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technicai Package of Poiicies, Programs, and Practices [11] to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments 167 168 169 170I 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 86 187 88 89 190 91 192 193 194 195 196 197r 198 199 200 201 202 203 204 205 206 207 208 209 2 i 0 h?s Rev 3.26.13, e-clearance The authors acknowledge Robert Anderson, Hollv Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, iDivision of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention, Web-hosed injury Statistics Query and Reporting System Atlanta, GA: National Center for injury Prevention and Control. Retrieved March 15, 2018. 2016. 2. lvey-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66113}: p. 1-16. 3. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, MD. 4. Kochanek, K., et al., Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Rosenrnan, 5.1., Preventing suicide: what will work and what will not. The Medical Journal of Australia, 1998. 159(2): p. 100-102. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention E??orts Unite to Address Rising National Suicide Rate. 2017: Washington, D.C. 8. Crepeau-Hobson, F., The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives ofSuicide Research, 2010. 14(1): p. 24*34. 9. American Association, Diagnostic and statistical manual of mental disorders 2013: American Pub. 10. Simon, T.R., et Characteristics of impulsive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51}: p. 4959. 11. Stone, D.M., et al., Preventing suicide: A technical package ofpoiicies, programs, and practice. 2017. 12. World Health Organization, Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO, 2012. 13. Caine, E.D., et aI., Comprehensive, integrated approaches to suicide prevention: practical guidance. injury prevention, 2017: p. injuryprev-2017-042366. 14. Milner, A., J. Sveticic, and D. De Leo, Suicide in the absence of mental disorder? A review of autopsy studies across countries. int Soc 2013. 59(6): p. 545-54. 15. Pouliot, L. and D. De Leo, Critical issues in autopsy studies. Suicide Life Threat Behav, 2006. 36(5): p. 491-510. Attachment: Suicide Vital Signs 3.26.18 lTables_Fig, e-clearance) Rev 3.26.18, e-clearance 213 214 Word Count: 177?1f1300 Ftev 3,2518, e-clearance Comment ltgs?l: I think we need a transition like this to avoid implying that the circumstances listed are the only ,significant differences. Short Title: Vital Signs: Contributing Circumstanc?s to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 2 3- lVlS? Kristin M. Holland, PhD;1Asha 2. Ivey-Stephenson, Alex E. Crosby, M01 Comment [tgsli]: You had changed this to .05. I don't think we need this in the abstract since we say they were signi?cant. Just saying this is consistent ,with the too. 4 Structured abstract [2451 250 words?this word count is not included in the 1300 max for the remainder] I i IE 5 Background: Suicide rates have risen nearly 30% since 1999. Mental health problems are just one factor 5' is contributing to suicide. Examining state?level trends in suicide and other contributing circumstances can inform i i u' i 7 comprehensive state suicide prevention planning. 8 Methods: Trends in age?adjusted suicide rates among people aged :10 years, by state and sex, across six 9 consecutive three-year periods [1999?2016], were assessed using data from the National Vital Statistics System It} for 50 states and Washington, DJ: Data from the National Violent Death Reporting System, covering 22 states in 2015, were used to examine contributing circumstances among decedents with and without known [2 IVIHP. l3 Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases Si IA of 30% or more. Rates increased signi?cantly among males and females, in 34 and 43 states, respectively. Over [5 half of decedents did not have a known MHP. Among decedents with circumstance information, several I [7 relationship problems/loss [45.1% vs life stressors [54.2% vs 49.2%] and recentfimpending crises [329% IS vs were significantly more likely among those without a knowo but l9 common across groups 20 Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances 2 contributed to suicides among people with and without known lleP. 22 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 23 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 24 after a suicide occurs. 35 INTRODUCTION 26 BACKGROUND AND PURPOSE 2? in 2016, nearly 45,000 suicides [lib/100,000 [age?adjustedll occurred in the United States among people 28 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 29 urbanization levels Suicide is the 10?? leading cause of death and is one ofjust three leading causes that 30 are increasing Additionally, rates of Emergency Department visits for nonfatal self-harma key risk factor 3 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 32 nation more than $69 billion in direct medical and work loss costs 33 The National Strategy for Suicide PreventioanSSP} calls for a public health approach to suicide prevention 34 with efforts spanning across multiple levels lie, individual, family/relationship, community, and societal}. Such 35 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 36 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing circumstances include social and economic problems, 38 access to lethal means leg, substances, firearms, bridges] among people at risk, poor coping and problem- 39 solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that Ftev 3.26.18, e-clearance 40 contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of Comment ?Est?: We can?t include this here because not all of the comparisons 4 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state?specific 42 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi-level i were Si?ni?c??t- I added SOME text to address this. Comment [tng]: I realized that we split up the results about methods and that our toxicology results are incomplete. suggest moving these down. 43 comprehensive suicide prevention. 44 METHODS 45 Suicide rates were analyzed for people aged :10 years only, as attributions of suicidal intent in younger children as are variable Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death 4? certificate records lintemotionoi Classrficotion of Diseases 10?? Revision, underlying-cause-of death codes X60- 48 K34, YEIO, Age-specific population estimates were obtained from US. Census Bureaquational Center for 49 Health Statistics bridged-race population data releases. 50 5 National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods 52 spanning 1999-2016. Rate estimates were age-adjusted to the U.5. year 2000 standard population and expressed 53 per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?year data 54 aggregates, employing weighted least squares regression with inverse?variance weighting. Modeled rate trends are reported in terms of average annual percentage changes 50 i 5? Characteristics [Table 2} and circumstances (Table 3} of suicide decedents 210 years, with and without known 58 MHP, were compared in the 2? states with complete data participating in National Violent Death 5' 59 Reporting System in 2015. defines l'v?lHP as disorders and listed in the Diagnostic and 60 Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance 6 dependence, which are captured separately in MUD-RS aggregates data from three primary data sources: 62 death Certificates, coronerfmedical examiner reports {including toxicology}, and law enforcement i 63 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression 64 analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and i 65 racer?ethnicity. as RESULTS 6? The most recent overall suicide rates (representing 2014-2010} varied four-fold, from 5.9 to 29.2 5 03 {Montana} per 100,000 persons per year {Table Across the study period, rates increased in all states, except i 69 Nevada [with the highest current suicide rate}, with absolute increases ranging from +0.3 {Delaware} to +3.1 5 70 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% [Delaware] to 67.6% (North Dakota}, Tl with increases more than 30% observed in 25 states [Table 1, Figure T2 I 7?3 Modeled suicide rate trends indicated significant increases for 44 states; for males [34 states} and females [43 T4 states}, as well as for the 0.5. overail (Table Nationally, the model?estimated MP0 for the overall suicide 75 rate was By sex, the national AAPC was significant for and femaIEs 76 feneales. T8 Suicide decedents without known MHP iN=11,030} were compared to those with iv'lHP 79 While all decedents were predominately male {Table 2; 80 26.8%} and non?Hispanic white those without known MHP, relative to those with MHP, were more 81 likely male (33.6% vs. 53.8%; ad?stedodds ratio 95% CI 2.2?2.5} and racialfethnic minorities {odds 82 ratio40R} range: 1.2?2.1; 95% CI range Suicide decedents without known MHP also had i 83 significantly greater odds of perpetrati 34 I I I 2 ?2.5.3 .i 1.311 .E if.? 133.5. 9.59-1! 1 lieffdaj; 100 101 (102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 12] 122 123 124 Rev 3.26.18, e-clearance Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. Altheangfirearms were the most common method overall usedrmeseefteereveealH43.5%) and for both groups, decedents witho_ut known MHP were more likely to die by firearm (55.3% vs. 40.6%) and less likely to Msuffocation (26.9% vs gandpoisoning (10.4% vs than those without known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test pesitive for any substance overall (a0R=0.8, 95% 0:07-08) but more likely to test positive for alcohol 95% All suicide decedents with keawa-MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MH P, versus those with known MHP had a signifieantly?greater likelihood (pd?.9Hof any relationship problem/loss (45.1% vs. specifically intimate partner problems (302% vs. arguments/conflicts (17.5% vs. and recently perpetrating interpersonal violence vs. They were also signifieantly-more likely to have experienced other life stressors (54.2% vs such as criminal- legal problems (10.7% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adiusted models. Among_hose with crises intimate partner and physical health problems were the most common types for both groups and did not differ between them. Wig?beth?gseeps?t-hemest respeetlvely. Decedents without known MHP had significantly lower odds of recent release (a0R=0.5 95% CI from any institution, but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MH P, were less likely to have a history of suicidal ideation 95% Cl=0.4v0.5) and prior suicide attempt 95% More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Additional research into the specific causes of these trends is necessary. Fortunately, data from the states participating in can shed light on the circumstances that contributed to recent suicides and help guide prevention activities. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known This group suffered more 3 125 126 127 128 129 131] 131 132 133 134 135 136 137 138 139 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 Rev 3.26.18, e-clearance relationship problems and life stressors such as criminal-legal matters, evictionlloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MHP, two?thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as jobifinancial, relationship, andior physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state?level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (Le. decreasing suicide rates), or decreased percentages of such deaths over time (Le. increasing suicide rates}. Second, is not yet nationally representative; the 27 states included in the current study represent 49.6% of the U.S. population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in?depth interviews with next?of?kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence unreported by key informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: 0 Technicai Package of Poiicies, Programs, and Practices [11] to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 .184 185 186 187 188 189 190 191 192 193 194 195 196 197r 198 199 200 201 202 203 204 205 206 207 208 209 210 211 Rev 3.26.13, e-clearance The authors acknowledge Robert Anderson, Hollv Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, iDivision of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atlanta, GA: National Center for injury Prevention and Control. Retrieved March 15, 2018. 2016. 2. Nev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 201?. 66113}: p. 1-16. 3. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, MD. 4. Kochanek, K., et al., Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Rosenrnan, 5.1., Preventing suicide: what will work and what will not. The Medical Journal of Australia, 1998. 15912}: p. 100-102. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention E??orts Unite to Address Rising National Suicide Rate. 2017: Washington, DC. 8. Crepeau-Hobson, F., The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives ofSuicide Research, 2010. 14(1): p. 2434. 9. American Association, Diagnostic and statistical manual of mental disorders 2013: American Pub. 10. Simon, T.R., et Characteristics of impulsive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51}: p. 4959. 11. Stone, D.M., et al., Preventing suicide: A technical package ofpoiicies, programs, and practice. 201?. 12. World Health Organization, Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO, 2012. 13. Caine, E.D., et aI., Comprehensive, integrated approaches to suicide prevention: practical guidance. injury prevention, 2017: p. injuryprev-2017-042366. 14. Milner, A., J. Sveticic, and D. De Leo, Suicide in the absence of mental disorder? A review autopsy studies across countries. int Soc 2013. 59(6): p. 545-54. 15. Pouliot, L. and D. De Leo, Critical issues in autopsy studies. Suicide Life Threat Behav, 2006. 3615}: p. 491-510. Attachment: Suicide Vital Signs 3.26.18 lTables_Fig, e-clearance) Rev 3.26.18, e-clearance 212 213 Word Count: 177?1f1300 Rev 3.26.18, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Structured abstract [2451250 words?this word count is not included in the 1800 max for the remainder] Background: Suicide rates have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods [1999-2016], were assessed using data from the National Vital Statistics System for 50 states and Washington, 0.0 (0.0.). Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known MHP. Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of 30% or more. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, those without known MHP were more likely (all p301) than those with a MHP to have relationship problemsfloss (45.1% vs life stressors [54.2% vs 49.7%) and recentfimpending crises (32.9% vs but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999?2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjusted]) occurred in the United States among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 101th leading cause of death and is one of just three leading causes that are increasing Additionally, rates of Emergency Department visits for nonfatal self-ha rm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and selfsharm injuries cost the nation more than $69 billion in direct medical and work less costs The National Strotegyfor Suicide calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal]. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi?determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing circumstances include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problem- solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of Rev 3.26.18, e-clearance reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 2:10 years only, as attributions of suicidal intent in younger children are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Ciassification of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, U03). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics (Table 2) and circumstances (Table 3) of suicide decedents 3210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racefethnicity. RESULTS The most recent overall suicide rates (representing 2014?2016) varied four-fold, from 6.9 (D.C.) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except Nevada (with the 9th highest current suicide rate), with absolute increases ranging from +0.8 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from (Delaware) to +57.6% (North Dakota), with increases more than 30% observed in 25 states (Table 1, Figure 1). Modeled suicide rate trends indicated significant increases for 444 states, for males (34 states) and females (43 states), as well as for the U.S. overall (Table 1). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, the national AAPC was for males and for females. Suicide decedents without known MHP were compared to those with MHP all chi-square and logistic regression analyses significant at p505). While all decedents were predominately male {Table 76.8%) and non?Hispanic white those without known MHP, relative to those with MHP, were more likely male (83.6% vs. 68.8%; adjusted odds ratio 95% CI 2.22.5} and racialiethnic minorities (odds ratio [08] range: 1.2-2.1; 95% Cl range Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide 2.9, 95% CI of firearm suicide (308 1.6, 95% CI and of testing positive for alcohol 1.2, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. 100 101 102 103 104 105 106 107 108 109 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 Rev 3.26.18, e-clearance Although firearms were used most often overall decedents with known MHP were more likely to die by suffocation (31.3 vs. 26.9%) and poisoning (19.8% vs. 10.4%} than those without known MH P. All suicide decedents with known MHP and approximately 85% without MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems 95% (2120.10.81. While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MH P, versus those with known MHP had a significantly greater likelihood of any relationship problemfloss (45.1% vs. specifically intimate partner problems (30.2% vs. (17.5% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other life stressors (54.2% vs such as criminal- legal problems (10.7% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. Among both groups, the most common crises were intimate partner (36.2% vs. 34.9%] and physical health problems (13.8% vs. respectively. Decedents without known MHP had significantly lower odds of recent release 95% Cl (040.5) from any institution, but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MH P, compared to those with MH P, were less likely to have a history of suicidal ideation 95% (2120405) and prior suicide attempt 95% More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Additional research into the specific causes of these trends is necessary. Fortunately, data from the states participating in can shed light on the circumstances that contributed to recent suicides and help guide prevention activities. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MH P, two?thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as job/?financial, relationship, andfor physical health problems. These findings point to the need to both prevent the conditions 3 126 127 128 129 130 131 132 133 134 I35 I36 .137 138 139 140 141 I42 I43 144 145 146 147 I48 149 l50 151 152 153 154 155 156 157 158 159 160 161 162 I63 164 165 Rev 3.26.13, e-clearance associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help?seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent decreasing suicide rates}, or decreased percentages of such deaths over time increasing suicide rates}. Second, is not yet nationally representative; the 2? states included in the current study represent 49.6% of the U.S. population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist it is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence unreported by key informants. However, any lack of awareness of decedent l'leP suggests, even further, the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technicai Package of Policies, Programs, and Practices (11] to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 77?0-488-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 3Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC Refe re nces: 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 Rev 3.26.18, e-clearance 1. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. 2. Ivev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. 3. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hvattsville, MD. 4. Kochanek, K., et al., Martaiity in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Genera.f and of the Nationai Action for Suicide Prevention. 2012. 6. Rosenman, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 169(2): 100-102. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 2017: Washington, D.C. 8. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicoi Examiners. Archives ofSuicide Research, 2010. 14(1): p. 2434. 9. American Association, Diagnostic and statisticai manuai ofmentai disorders 2013: American Pub. 10. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life Threatening Behavior, 2002. 32(51}: 49?59. 11. Stone, D.M., et al., Preventing suicide: A technicai package of poiicies, programs, and practice. 2017. 12. World Health Organization, Risks to mentai heaith: An overview of and risk factors. Geneva: WHO, 2012. 13. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicoi guidance. Injurv prevention, 2017: p. injurvprev?2017-042366. 14. Milner, A, J. Sveticic, and D. De Leo, Suicide in the absence of mentai disorder? A review of autopsy studies across countries. IntJ Soc 2013. 59(6): p. 545-54. 15. Pouliot, L. and D. De Leo, Criticai issues in autopsy studies. Suicide Life Threat Behav, 2006. 36(5): p. 491610. Attachment: Suicide Vital Signs 3.26.18 (Tables_Fig, e-clearance) Word Count: 177171800 Table 1. Trends in Suicide Rates among Persons :2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2991 2992 2994 2995 2997 2998 2919 2911 2913 2914 2916 Rank (State Rank) '9 (State Ragnk) H. Both 12.3 (We) 12.7 9.4) 12.9 9.2) 13.8 9.9) 14.5 9.8) 15.4 9.9) 1.5 ?70 (pr-1.91) nia 3.1 (nia) 25.4 (his) U.S. Male 29.9 (nia) 21.2 9.4) 21.3 9.9) 22.5 23.5 1.9) 24.5 1.9) Female 4.7 5.9 9.3) 5.3 9.2) 5.7 9.4) 6.2 9.5) 6.9 9.7) 2.6 31. Both 14.3 (nia) 13.4 (- 9.9) 14.1 9.6) 15.6 1.6) 16.4 9.7) 17.5 .1) +1.6 (ps. 95) 25 3.1 (31) 21.9 (33) AL Male 25.1 (nia) 23.4 (- 1.7) 24.4 1.9) 26.4 2.9) 27.6 1.1) 29.1 1. 5) 1. 3 (ps. 95) Female 5.1 (nia) 4.8 (- 9.3) 5.9 9.2) 1 6.4 9.3) 9. 7) 2. 6 (.ps 91) Both 21.9 (nia) 24.8 3.8) 24.2 (- 9.6) 26.9 1.7) 25.4 (- 9.5) 28.8 3.4) +1 7 (p4. 95) 2 7.8 4) 37.4 (13) AK Male 33.2 (nia) 38.1 4.9) 38.9 9.8) 49.1 49.1 (- 9.1) 42. 9 2. 8) +1.4 (ps. 91) Female 8.6 (nia) 11.4 2.9) 9.8 (- 1.6) 11.1 9.9 (- 1.his Both 17.8 (nia) 18.5 9.7) 19.1 9.5) 19.1(- 9.9) 29.4 1.3) 29.9 9. 5) +1.9 91) 15 3.1 (32) 17.3 (42) .42 Male 29.3 (nia) 39.2 1.9) 39. 6 9.4) 39. 2 9.5) 32. 9 1.(ps. 95) Female 7.1 (nia) 7.5 9.4) 9.7) 9.5) 9.6) 9. 6) 2. 2 (ps. 91) Both 15.5 (nia) 15.8 9.3) 16.2 9.5) 17.6(+ 1.4) 19.2 1.6) 21 91) 12 +5.7(14) AR Male 26 7 (nia) 26.7 9.9) 27. 2 9.5) 28. 2 1.9) 31. 7 3.5) 33. 5 1. 9) 1.6 (ps. 95) Female 5.6 (nia) 5.9 9.3) 6. 2 9 4) 1.7) 7.5 (- 9.4) 2.1) 3. 6 (.ps 91) Both 19.6 (nia) 11.3 11 93) 11.8 (- 9.1) 12.1 95) 45 1.6 (46) CA Male 17.9 (nia) 18.4 9.5) 17.7-( 9.7) 19.1 18.9 (-9.2) +95% his Female 4.1 (nia) 5.9 9.9) 4.9 9.1) 9.5) 5. 3 (- 9.1) 9. 3) 7 (ps. 95) Both 17.3 (nia) 19. 9 .2) 29. 9 1.9) 21.6 1.5) 23. 2 16) 1.8 (p4. 91) 8 5.9 (12) 34.1 (22) CO Male 28.6 (nia) 39.9 2.3) 39.5 (9 .4) 31.5 1.9) 33.4 1.9) 36. 3 2. 9) +1.4 (ps. 91) Female 7.9 (nia) 8.2 1.3) 8.2 9.(ps. 91) Both 9.6 (nia) 8.9 (w 9.7) 9.1 9.2) 19. 2 1.1) 11.9 9.8) 11.5 9.5) +1.6 (cs. 95) 46 1.9 (43) +192 (34) CT Male 16.4 (nia) 14.6 (w 1.8) 15.9 9.4) 16. 6 1.6) 17. 6 1.9) 17.3 (- 9.3) 9.9 his Female 3.6 (nia) 3.8 9.2) 3.7 (- 9.2) 9.7) 9.5) 6.2 1.3) 3.5 Rates are age-adjusted to the US. year 2999 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase to largest percentage decrease (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 29 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 1090 - 2001 2002 2004 2005 2007 2000 2010 2011 2013 2014 2015 Rank 5 (State Rank} 1' (State Ragnk} Both 15.0 (nfa} 1221+ 1.41 12.01? 0.01 10.0 1+ 1.0} 15.0 1+ 0.0} 10.0 1+ 0.2} 0 0 (prism) 02 0.0 (00} 30.0 111 (50} Male 20.0 {nits} 2001+ 2.151 20.01- 0.4} 23.5 1+ 2.0} 2001+ 0.51 2551+ 0.01 0 0 (111505} Female 5.3 {Na} 0211+ 0211 001+ 0.51 5.0 1+ 0.0} 0.4 1+ 0.5} 021+ 0.51 3.0 (prism) Beth 15.0 (nfa) 15.4 1+ 0.0} 1531+ 0.01 115.3 1+ 0.51 1001+ 12.01 10.0 1+ 0.0} 0.0 (pst?i) 50 0.0 (40} 46.0 ?10 (45} as Male 23.71011} 25.0 1+ 9.3} 3901+ 1101 3501+ 1151 33.1.11 3.51 111.71+ 1.5) 11.9 11, (1111501 1 Female 1100111151111 2.0011000? 531+ 0.01 5.01- 0.3) 0.0 1+ 1.0} 001+ 0.01 3 0 (111505} Both 14.0 {Ma} 15.0 1+ 0.0} 1001+ 0.31 10.3! 1+ 0.01 1021+ 0.01 10.0 1+ 0.1} 0. 0 11:. 1p<. 05 20 5.0 140} 50.0 '10 (40} E11 Male 25.0 {Ma} 20.0 1+ 0.0} 2001+ 0.01 2021+ 201 $131+ 0.01 2001+ 0-01 0. 5 (01} Female 5.0 (nfa} 0.0 1+ 0.5} 0.0 1+ 0.0} 0.1 1+ 0.5} 1.0 1+ 0.0} 7.0 1+ 0.0} 5.0 1p<.01} Both 13.9 {nra} 1301+ 0.31 1301+ 0.91 13.9 1+ 0.9} 10.3 1+ 0.3} 13.0 1+ 2.3) 11.9 11. 1111505) 39 3.3 133) 39.3 11. 130} GA Male 22.0 We} 2201+ 0.01 2101+ 0.01 25.0 1+ 0.0} 20.0 1+ 0.5} 2001+ 5.0} 0.5 n13 Female 5.0 {Na} 4.01- 0.2} 051+ 0151 001+ 0.01 0.5 1+ 0.2} 051+ 0.0} 2.0 1p<. 05} Both 12.01nta} 13.01? 0.0} 1001+ 0.01 15.5 1+ 4.0} 1001+ 0.51 1001+ 0.01 2.0 (111505} 05 0.0 135} 20.0 20 (30} 101E Male 20.0 {nits} 22.0 0.1} 0001+ 1.91 2501+ 5.1} 02.5 1+ 0.5} 2001+ 1.51 2.0 '11. (111505} Female 5.0 {Na} 5.0 0.0} 551+ 0.51 0.0 1+ 0.5} 5.0 1+ 0.01 1101+ 0.31 3.2 (111505 Both 10.0 (nta) 10.0 1+ 0.0} 1001- 0.0} 20.0 1+ 0.3} 20.0 1+ 0.5} 20.0 1+ 0.0} 0.5 '11: 1p<. 05} 40 5125114030} 4110352001410 01} 001} Male 20.0 (nfa} 50.0 1+ 0.1} 331.31- 0.0} 50.0 1+ 0.0} m11+ 0:51 3001+ 031 0.0 (p150 5} Female 13.0 {nits} 0.0 1+ 0.111 0.0 1+ 0.0} 00' 1+ 201 0.5 1+ 0.5} 1401+ 0.0} 4.0 '11. 1p<. 05} Both 0.0 {hrs} 001+ 0.121 0.4? 1+ 0.01 10.0 1+ 0.0} 10.0 1+ 0.5} 10.0 1+ 0.0} 2.5 11:. 1p<. 05} 40 ++220211100111 03023810112020} MIA Male 12.1 {Ma} 1001+ 0.51 1021+ 0.51 15.5 1+ 2.4} 1051+ 0.01 10.0 1+ 0.0} 2.0 1p<. 05} Female 3.0 (nfa} 201 0.0} 0.0 1+ 0.0} 0211+ 0.111 4.0 1+ 0.0} 001+ 0.51 3.0 1p<. 05} Both 13.9103} 13.5 1+ 0.1} 13.9 1+ 0.1} 1391+ 0.0} 1001+ as) 13.0 1+ 11.1} 1.9 11. 1p<. 01} 33 3.9126) 33.9 11.129} itll Male 20.0 We} 20.0 1+ 0.0} 24.5 1+ 0.2} 20.0 1+ 0.0} 20.0 1+ 2.0} 20.0 1+ 1.0} 1.5 1p<. 01} Female 4.5 {Na} 5.0 1+ 0.4} 5.0 1+ 0.2} 5.0 1+ 0.0} 0.0 1+ 0.0} 0151+ 0.01 2.0 1p<. 01} Rates are age-adjusted to the U.S. year 2000 standard. 1 Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not signi?cant. Current state rank 150 states and the District of Columbia} is for the reporting period 2014 .. 2010. Ranks are from highest rate to lowest rate 1511. ranks do not necessarily imply a statistically signi?cant difference. 1' Overall rate change is between the first (1000 2001} and last (2014 2016} reporting periods. Ranks are from largest increase to largest decrease Differences between ranks do not necessarily imply a statistically signi?cant difference. Overall percent change in rates is between the ?rst (1000 - 2001 and last (2014 2016} reporting periods. Ranks are from largest percentage increase to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 0 Rate based on 20 suicides. Differences between Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last pen'ods and might have contributed to lower reported rates. 1? Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 11.-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 (We) 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 41n1a1 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 (Ma) 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.51n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20151 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51n7a) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30 Till) 34.1 ?fa (23 RI Male 15.4 (nfa) .2) 14.8(- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.16.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (nfa) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (nfa) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18.2 1.0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 26.8 1.3) 8 0 1. 2) 28.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (nfa) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (nfa) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3 1W) 46.5 3?a( 4 UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (nfa) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2(+4 0. 9) 6.4 1.3) 6 6 0.2) 7. 3 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (nfa) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 25.1 1+1 .0) 25. a 1+ 1:1. 9) 27.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nl?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 Decrease 1.0% I:I ncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 37.8% - 57.8% Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states", 2015 Characteristics Total Known Mental No Known Chi- 0R Adjusted OR 5 (n=20,446) Health Mental Square {95% Cl) (95% Cl) Problem* Health ln=9,4D7) Problem {n=11,039) Sex Male 15,702l76.8) 6,469l68.8) 9,233l83.6) p<.01 Female 4,744l23.2) 1,806l16.4) p<.01 Age) 10-24 2,804l13.7) 1,593l14.4) p<.01 25-44 6,456l31.6) 3,036l32.3) 3,420l31.0) p<.05 DBMS-1.0) 45?64 3398853) p<.01 65+ 2,128l193) peol Racefethnicitv White, non-Hispanic 17,102l83.6) 8,165l86.8) 8,937l81.0) p<.01 Black, non~Hispanic 411(4.4) 8170.4) p<.01 American Indiaanlaska Native, non~ Hispanic 378(1.3) 112(1.2) 2660.4) p<.01 Asian, non?Hispanic 235(2.5) 341(3.1] p<_05 Hispanic 1,096l5.4) 463(43) 6336.?) p<.05 Other 66l0.3) 21(0.2) 4500.4) p<.05 Extended demographics Ever served in military? 3,429l173) 213750.01) p<.01 Homeless 240l1.2) 104(1.l) 136(13) Incident Type Single suicide 20,063l98.2) 9,318l99.1) p<.01 Homicide followed by suicide 319(1.6) 64(0.7) 255(2.3) p<.01 3.5{2.6r4.5) Multiple suicide 64l0.3] 39(0.4) Method Firearm 9,909l485) 3,821l40.6) 613881553) p<.01 2,940l313) Poisoning 3,003l14.7) 1,861l19.8) p<.01 Substance class causing death?r Other over-the-counter) 1,021l34.0) 66685.8) 355(31.1) pc?l Opioids Antidepressants Benzodiazepines Antipsvehoties Other Toxicology Results Any toxicology testing Positive for a 1 substance? Substance detected? Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive 944(314) 800126.63) 6241208) 2190.3) 1.595(13) 9,913l74.4l 3.554(413) 2.229995) 3.12439?) 2.454(303) 19788933) 499(53) 7.515(372) rat-star) 5.559(321) 1.471(224) 5,425l265l 503(327) 644(346) 468(25.1] 195(105) 239(33) 6,658l70.8) 5,192l78.0) 4,253t4s.3) 4,225t449) 2155.6) 16951393) 3r6(1o.2) 3.127(332) 21092.7) 335(294) 155(13.7) 155(13.7) 24(2.1) 3150.4) 6,659l603) 4,721l70.9) 2.327(420) 4.295(339) 3,898l353) 325(21.2) 283(6.9] 33191395) 360(9.2} ?61(22.1) 2,322l21.0) 479(20.5) p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 aster-1.0) arias-0.7) Dates-1.0) totes-1.1) news-1.0) earns-0.5) routs-1.1) Alaska, Arizona, 'Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts. Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma. Oregon, Rhode Island, South lCarolina, Utah, Vermont, T?v?irginia, and Wisconsin. 1' had been identi?ed as having a current diagnosis of mental health problem in eoronerr?medieal examiner or law enforcement reports. Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex. race and ethnicity. were aged 10 years and older, as per standard in the suicide prevention literature. M?D'enominator is aged 18 years of age and older with reported military service status. TT Denominator is decedents who died by poisoning, including overdose. ii Denominator is decedents with any toxicology tested. 1" Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states?, 2015 Characteristics Total Known Mental No Known Chi- 0R Adjusted 0R5 Health Mental Square {95% Cl) {95% Cl} Problem" Heath Problem Suicide with known circumstances 93571848} p<.01 Mental HealthSubstance Problems Any Current Mental Health Diagnosis? 7,075l752} Anxiety disorder 1,579i16.8} Bipolar disorder 1,431i15.2} Schizophrenia 5095.4) 509(5.4) PTSD 424(45} 226{2.4) 226(2.4) Unknown 760{8.1) ?60(8.1) Current depressed mood 10388715} 3,0?6i323} p<.01 Substance Problems Any Current substance problem 2,343i25.0} p<.01 0.710.103) Alcohol problem 3,268i17.4} 1,406i15?} p<.01 Other substance problem 3,084i16.4} 1,?68i183] p<.01 Treatment Current mental health/substance abuse treatment p<.01 Ever treated for metnal problem p<.01 Relationship Problemslloss Any relationship problemfloss 7,948l42.4) p<.01 Intimate partner problem 5,093i27.2} 2.828(302} p<.01 1.4{1.341.5) Perpetrator of interpersonal violence past month 414{2.2) 131(1.4) 2838.0) p<.01 Victim of interpersonal violence within past month 84{0.4} 31(03) p<.05 Family relationship problem 798(8.5) Other relationship problem [non-intimate) 403{2.1] 202(2.1) 201(2.1) Argument or conflict {not specified} p<.01 Death of a loved one {any} 1,497i8?) 826(38): 671(7?2) p<.01 Non-suicide death 1,181i5.3] 5345.?) p<.01 Suicide of family or friend Other Life Stressors Any life stressor Recent criminal legal problem Other legal problem Physical health problem Jobeinancial problemM Eviction or loss of home School problem?r1r Recent release from an institution?? Jailfprison/detention facility Hospital hospital/institution Other (includes alcfSA treatment facilities} Recent or Impending Crisis Crisis within past or upcoming two weeks? Intimate partner problem crisis Physical health problem crisis Criminal legal problem crisis Family relationship problem crisis Job problem crisis Suicide EventlHistory Left a note Disclosed suicide intent History of ideation History of attempts 37912.0) 9,743151.9) 1533135) ?143140) 4,1791223) 29411102) 722133) 152119.31 1,41217.s) 203114.41 517133.13) 459133.21 2231153) 55251294) 19531355) 31391134) 3211112) 43017.3) 35415.4) 5,45313451 44051235) 5390813} 30321190} 21712.3) 4.3751407) 53513.2) 33140) 20121214) 15301103) 31713.4) 70117.3) 941110.21 3213.?) 311133.01 439146.71 109111.51 2,444l26.0} 354134.91 315112.91 20313.3) 21213.?) 19117.3) 3.1321333) 23031245) 33331403) 23701294) 15211.?) 5.0331542) 1.0021107) 37014.0) 2,157123.2) 1411i15.6} 40514.3) 92121.9) 47115.1) 121125.?) 206l43.7) 3016.4) 114124.21 30311319) 11141332) 424113.31 413113.15) 21317.1) 16315.3) 3,2351351) 2,0991224) 11521230} 9621103) p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 0.710.609) 1.211.103) 1.311.500) 1.010.301) 1.111.012) 09103-10) 1.311.1-1.5) 1.310.3-13) 0.510.405) 3.612.149) 1.311.300) 0.110.101) 2.411.3-33) 1.411.305) 11109-12) 11109-13) 1.711.501) 0.3100210) 0.110.503) 1.111.011) 09103-10) 0.410.405) 0.310.303) 0.310.700) 1.111.1-1.2) 1.711.509) 1.010.942) 1.011.011) 0.910.310) 1.411.2-1.5) 13109-13) 0.510.405) 4.513.204) 1.311.001) 0.110.101) 2.511.333) 1.411.345) 1.110.942) 1.010.312) 1.511.303) 0.910.111) 0.710.503) 1.211.1-1.2) 0.910.309) 0.410.405) 0.310.303) *Alaska, Arizona. Colorado, Connecticut, Georgia. Hawaii, Kansas. Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey. New Mexico, New York, North Carolina, Ohio, Oklahoma, ISiregoin, Rhode island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Dceedcnt had been identified as having a current diagnosis of mental health problem in examiner or law enforcement reports. 5? Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age. sex, race and ethnicity. ?Denominator is decedents aged [8 years of age and older. 11' Denominator is decedents aged 10-18 years. ii Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month. 1'5 Denominator of crisis subgroup is decedents with any crisis within past or upeomin two weeks. Crises depicted here represent the most commonly oecurrin categories. Table 1. Trends in Suicide Rates among Persons 1: 19 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 109,999 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2091 2002 2904 2095 2007 2903 2010 2011 2913 2914 2015 Rank (State Rank) 11 (State Ragnk) H. Both 12.3 (We) 1271+ 0.4) 12.9 0.2) 13.31+ 0.9) 14.5 1+ 0.3) 1541+ 0.9) 1.5 c111154.01) nia 3.1 (Na) 25.4 a. (nia) U.S. Male 29.9 (nia) 2121+ 0.4) 2131+ 0.0) 22.5 23.51+1.0) 24.51+1.0) Female 4.7 5.0 1+ 0.3) 5.3 1+ 0.2) 5.7 1+ 0.4) 5.2 1+ 0.5) 5.9 1+ 0.7) 2.5 %1p<.01) Both 14.3 (nia) 13.41- 0.9) 14.1 1+ 0.5) 15.51+1.5) 15.41+ 0.7) 17. 51+ .1) +1.5 %1p<.05) 25 3.1131) 21.9 %133) AL Male 25.1 (nia) 23.4 (- 1.7) 24.41+ 1.0) 25.41+ 2.0) 27.5 1+ 1.1) 29.1 1+ 1. 5) 1. 3 ?/61 (cs. 05) Female 5.1 (nia) 4.31- 0.3) 501+ 0.2) 1 5.41+ 0.3) .01+ 0. 7) 2. 5 5'41 01) Both 21.0 (nia) 2431+ 3.3) 24.21- 0.5) 2501+ 1.7) 25.41- 0.5) 2331+ 3.4) +1 7 (psi 05) 2 7.31 4) 37.4 113) AK Male 33.2 (nia) 3311+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 0.1) 42. 9 1+ 2. 3) +1.4 1p<. 01) Female 3.51nia) 11.41+ 2.9) 9.31-1.5) 11.1 1+ 1.2) 991-12nis Both 17.31nia) 1351+ 0.7) 19.1 1+ 0.5) 19.11- 0.0) 20. 4 1+ 1.3) 20. 9 1+ 0. 5) +1.0 %11.p< 01) 15 3.1132) 17.3 142) .42 Male 29.3 (nia) 3021+ 1.0) 30. 5 1+ 0.4) 30. 2 1? 0.5) 32. 0 1+ 1.1pc. 05) Female 7.1 (nia) 7.51+ 0.4) 21+ 0.7) 51+ 0.5) 21+ 0.5) 91+ 0. 5) 2. 2 1p<. 01) Both 15.5 We) 1531+ 0.3) 15.21+0.5) 1751+ 1.4) 1921+ 1.5) 21 g..21+20) 01) 12 +5.7114) +35.3%115) AR Male 25 71nia) 2571+ 0.0) 27. 2 1+ 0.5) 23. 2 1+ 1.0) 31. 7 1+ 3.5) 33. 5 1+ 1. 9) 1.5 5'41 1pc. 05) Female 5.5 (nia) 5.91+ 0.3) 5. 2 1+ 0 4) 91+ 1.7) 7.51- 0.4) 51+ 2.1) 3. 5 31:1 01) Both 10.5 (nia) 11.3 11 .01- 03) 12.01+1.0) 11.31- 0.1) 121 05) 45 1.5145) +14.3%145) CA Male 17.9 (nia) 1341+ 0.5) 17.7-( l0.7) 19.1 1+ 1.4) 13.9 (-0.2) 19.21+0.3) +05% nis Female 4.1 (nia) 501+ 09) 4. 0. 1) 41+ 0.5) 5. 3 1- 0.1) .51+ 0. 3) 7 5'41 1pc. 05) Both 17.31nia) 19.21+1.9) 19. 01-0 2) 20. 0 1+ 1.0) 21.51+1.5) 2321+ 15) 1.3 (psi 01) 3 5.9112) 34.1 122) CO Male 23.5 (nia) 3091+ 2.3) 30.510 .4) 31.51+1.0) 33.41+1.9) 35. 3 1+ 2. 9) +1.4 1p<. 01) Female 7.0 (nia) 3.21+1.3) 321+ 0.0) 11+ 0.9) 10.1 1+ 1.?/61 (cs. 01) Both 9.5 (nia) 3.91? 0.7) 9.1 1+ 0.2) 10. 2 1+ 1.1) 11.01+ 0.3) 1151+ 0.5) +1.5 (cs. 05) 45 1.9143) +192 11211134) CT Male 15.4 (nia) 14.5 1? 1.3) 1501+ 0.4) 15. 5 1+ 1.5) 17. 5 1+ 1.0) 17.31- 0.3) 0.9 his Female 3.5 (nia) 331+ 0.2) 3.71- 0.2) 41+ 0.7) 91+ 0.5) 5.21+ 1.3) 3.5 %1p<.05) Rates are age-adjusted to the US. year 2000 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest percentage increase 11} to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 20 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank} 1? (State REnk} Both 13.6 1n1a1 12.21.41 1191? 0.31 13.61+ 1.71 14.21+ 0.61 14.41+ 0.21 0.9 nis 42 0.81501 5.9 ?it: 1501 DE Male 23. 0 1n1a1 20.31.71 1991- 0.41 23.1 1+ 3.21 22. 71- 0. 41 2351+ 0.81 0.6 ?fa Female 31ni'a1 010 .21 4.61? 0.41 91+ 0.31 6. 41 1.51 21- 0.21 1.6 nl's Both 91nia1 6.41+ 0.51 6.41? 0.01 31+ 0.81 6?01.71 6.91+ 0.31 0.9 ?fa 51 1.01481 16.1 ?it 1451 DC Male 10.71n1a1 11.11+ 0.41 10.31; 0.81 12.71+ 2.41 10. 012 .61 11. 71+ 1.71 0.3 ?it: nis Female 1.71nia1?FT 2.31+ 0.611?r 331+ 1.01 61- 0.71 3.61+ 1018-0181 3.5 ?fa Both 14.81n1a1 1521+ 0.41 14.91- 0.31 16. 3 1+ 1.41 16. 31?0 .1 16.41+ 0.11 0.8 301p<.051 29 1.61451 10.6 i131481 FL Male 24.3 1n1a1 24.41+ 0.11 23.61- 0.81 26. 2 1+ 2.61 25. 61 0. 61 25.6 1?0. 11 0.5 ?it. Ms Female 6 3 (ma) 681+ 0.51 6.81+ 0.0111+ 0.31 7.61+ 0.51 81+ 0.31 1.4 301p<011 Both 12.91n1a1 13.21+ 0.31 12.31? 0.91 13. 2 1+ 0.91 13.71+ 0.51 1501+ 1.31 0.9 ?it; nits 39 2.1 1401 +162 131441 GA Male 22.1 1n1a1 23.1 1+ 1.01 21311.81 .91+ 0.61 2261+ 0.71 24.-41+ 1.71 0.5 ?fa nl's Female 5.0 1n1a1810.21 4.6 1- 0.2151+ 0.91 5.8 1+ 0.31 6.6 1+ 0.81 2.1 ?fa 1p<.051 Both 12.91nla1 11. 1 11.81 10.31- 0.71 14. 5 1+ 4.11 14.41? 0.11 1521+ 0.81 2.0 ?fa 35 2.41351 18.3 I1111381 HI Male 20.4 1n1a1 17.21.11 15.3 1- 1.91 21. 91+ 6.71 2251+ 0.51 24.31+ 1.81 2.1 ?fa Female 5.4 1n1a1010.41 551+ 0.5111+ 1.51 21- 0 91 5.91- 0.31 1.2 ?fa nl's Both 17.31n1a1 1921+ 2.01 18.31? 0.91 21. 61+ 3.31 2191+ 0.31 24.71+ 2.81 2.3 3611:1101} 6 7.51 61 43.2 1 71 ID Male 28.4 1n1a1 33.1 1+ 4.71 31.1 1? 2.01 3 91+ 3.81 71- 0 21 3801+ 3.31 1.6 301p<051 Female 7.2 1n1a1 6.1 1- 1.11 6.1 1+ 0.0101+ 2.91 9.51+ 0.51 11.81+ 2.31 4.4 1p<.051 Both 9.9 1n1a1 9.81? 0.11 9.71- 0.11 10. 6 1+ 0.81 11.21+ 0.61 1221+ 1.01 1.5 3131p<.051 44 2.31381 22.8 ?it 1321 IL Male 17.1 1n1a1 16.71? 0.41 16.2 1- 0.41 17. 61+ 1.41 18.51+ 0.91 1981+ 1.31 1.1 36113105) Female 3.7 1n1a1 3.814 0.01 381+ 0.2121+ 0.41 4.51+ 0.41 521+ 0.61 2.4 301p<011 Both 13.01n1a1 13.71+ 0.71 14.41+ 0. 71 1491+ 0.51 16.41+1.41 17.11+ 0.71 +1.9 31: 1p<.011 26 4.11231 31.9 13125} IN Male 22.4 1n1a1 23.21+ 0.81 24. 41+ 1.21 24. 71+ 0.41 2671+ 2.01 28.31+ 1.61 1.5 36113101} Female 4.6 1n1a1 501+ 0.41 531+ 0.2191+ 0.61 681+ 0.91 6.61? 0.21 2.7 361131011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia1 is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 (-1.4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nia) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 i1110.14.01) 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nia) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nia) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia.4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 ?it; 27 3.8 29.3 14(26) LA Male 22.9 (nia) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa Female 4.8 (nia) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 31;? Both 14.5 (nla) 1310) 18.9 3.5) 18. 0.4) 2.2 51: 21 4.0 (25) 2?.4 ?11} (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nia) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nia) 10.3 0.3) 10.1 0. 2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(02) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nia) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 90(4 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (nia) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nia) 4.8 0.0.9) 2.8 31;? Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State ank) Both 10.71n1a) 11.51+ 0.9) 12.41+ 0.5) 12.91+ 0.5) 14.21+1.3) 15. 01 0. 9) 2.3 701p<.01) 35 4.3119) 40.5 fit-r1 5) MN Male 15.3 1nia) 1931+ 1.1) 20.41+ 1.0) 2091+ 0.5) 22.9 81+ 1.9) 2331+ 0.4) 1.7 ?fa 1p<.01) Female 3.5 1n1a) 4.21+ 0.5) 51+ 0.5) 1 1+ 0.4) 05) 591+ 1.2) 4.2 55113101) Both 12.91nia) 14.1 1+ 1.2) 14. 7 1+ 0. 5) 1551+ 0.5) 1551+ 0.1) 15. 210 .3) +1.1 3511:1105) 35 2.3135) 17.8 i55140) M5 Male 22.9 1n1a) 24.5 1+ 1.7) 25.1 1+ 0 5) 2551+ 1.7) 25.910 .9) 25. 31.5) 0.7 nlis Female 4.3 1nia) 501+ 0.541+ 0.9) 210.2) 2.4 1p<.01) Both 14.71n1a) 14.1 1? 0 5) 15.41+1.3) 1501+ 0. 7) 17.51+ 1.7) 20. 0 1+ 2.3) 2.2 15 5.3115) 35.4 i15117) MD Male 25.3 1nia) 23.71? 1.5) 25. 5 1+ 1 .9) 2551+ 1.0) 2591+ 2.3) 32. 21+ 3.3) 1.5 501p<.05) Female 5 41n1a) 5.41+ 0.1) 11+ 0. 7) 31+ 0.2) 7.41+ 1.1) 51+ 1.2) 3.2 501p<.01) Both 21.1 1nia) 2251+ 1.4) 23. 5 1+ 1 .0) 24.71+1.1) 2571+ 2.0) 29.21+ 2.5) 2.1 1 5.01 2) 35.0 35111) MT Male 35.9 1n7a) 3731+ 0.0.1) 41 0.1+ 1 .4) 4551+ 4.4) 1.3 55113101) Female 5.71nia) 5.41+1.5) 41 0.1) 10. 0 1+ 1. 5) 12.51+ 2.5) 1311+ 0.5) 4.5 3511:1101) Both 12.71ni'a) 1221? 0 5) 12. 51+ 0 .4) 11. 71?0 5) 1351+ 1.5) 14.51+1.3) +1.0 ?fa 40 2.1 142) 15.2 i111143) NE Male 22.21nia) 2071-1523.91+19.) 0.5 ?fa nis Female 3.5 1n7a) 4.21+ 0.4)11+ 0 9) 01 1.2.5 ?70 his Both 23.31nia) 22.51- 0 5) 22.1 1 0. 5) 2251+ 0. 5) 21.4 2) 23.1 0.2 ?it. n15 9 0.2151) -1.0 55151) NV Male 35.3 1n1a) 35.714 1.7) 35.1 1 1 .5) 35510.7 51; 1115 Female 5.91nia) 951+ 0.5) 51+ 0.1) 100111.21+ 0. 5) +1.5 ?fa 1p<.01) Both 13.51n1a) 12.51?1.) 13.31+0.5) 15.21 1. 9) 15.51+0.5) 20.0 +2.7 3131p<.05) 17 +5.51 5) +453 %1 3) NH Male 22.51n1a) 21.1 11.4) 21.71+ 0. 5) 24. 5 1+ 3.2.2 5511:1105) Female 5.3 1n1a) 510.5) 91+ 1.3.9 501p<.05) Both 7.5 1n1a)7?01 1) 51 0.2 01+ 0.551.5147) 19.2 35135) NJ Male 13.01n1a) 13.11+ 0.0) 12.51 0.55) 13. 71+ 1.1 14.51+ 0.5) 1451+ 0.1) 0.9 %1p<.05) Female 3.2 1nia) 91? 0.3) 301+ 0.0) 91 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2006 2010 2011 - 2013 2014 2016 Rank 5 {State Rank) 17 (State REnk} 55m 22.01n1a1 2201- 011510.21 2301+ 1.21 24.1 1+ 1.11 25.01+1.91 +1.1351p<.051 4 4.01241 15.3 351391 NM Male 36.6 1nfa1 3737.1 1+ 1 .31 4071+ 3.61 0.4 35 nis Female 5.5 1n1a1411.1121011.71+ 0. 91 1201+ 0.31 3.3 351114.051 Both 7.2 1n1a1 110 .1+1. 11 310 .11 21351114011 49 2.1 1411 25.5 351271 NY Male 12.51n1a1 12.210.31 1291+ 0.71 1391+ 1. 01 1541+ 1.41 14.510 .51 1.4 351p<.051 Female 2.71nfa161-0.11 301+ 0311+ 0. 51 4.2 1+0. 71 461+ 0.51 +4.2 35 1p<.011 55m 13.61nl'a1 13510 .11 1371+ 0.11 14.21+0.51 14.5.1+041 15..31+051 +0.5 35154011 34 1.71441 12.7 35 1471 no Male 22.7 1n1a1 2271+ 0.01 22.2 1- 0.51 2331+ 1.11 2331n75 Female 5 61nfa1 51- 0 21 621+ 0.2.0 35113905} Both 13.31nl'a1 14.61+1.31 1601+ 1.41 16. 6 1+ 0. 61 1641+ 1.91 20. 9 1+ 2. 51 2.9 351p<.011 14 7.61 51 57.6 351 11 no Male 21.4 1n1a1 2451+ 3.21 23012.5 351114.011 Female 5.5 1n1a1 51- 1.01 371- 0.5171+ 2 01 5. 7 1+ 1.01 551+ 1 5.1 3.9 35 n75 Both 11.5 1n1a1 1231+ 0.51 131 1+ 0.51 1341+ 0.21 14.51+ 1.41 15.51+1.01 +2.0 35 1p<.011 32 +4.21211 +350 351191 OH Male 20.4 1n1a1 2091+ 0.51 2221+ 1.31 221p<.011 Female 4.0 1n1a1 471+ 0.71 491+ 0.1131+ 0.51091 5. 7 1+ 0. 51 3.4 351114.011 Both 17.0 1n1'a1 16. 511641+ 1.11 20. 7 1+ 2 31 2351+ 2.61 2.3 35109051 7 6.41101 37.6 i151121 OK Male 26.5 1nfa1 27.31.21 2761+ 0. 51 3031+ 2.51 3341+ 3. 11 3731+ 3.61 2.0 351139051 Female 6.6 1nfa1401- .21 7. 5 1+ 1.11 01- 0.51 651+ 1.61 10.31+ 1.61 2.9 1p<.051 Both 16.41nta1 17.71+1.31 17.71.01 16. 61+ 0.91 19. 6 1+ 1 .21 21.1 {+1.31 +1.6 351p<.011 13 4.61161 26.2 i551261 OR Male 27.41nl'a1 2951+ 2.11 25.51 0. 91 29.51+1.01 31.41+1.51 33 01+ 1.51 +1.1 351154.011 Female 6.51nfa1 1 1+ 0.61 7.71+ 0.6141+ 0.71 6612.7 35109011 Both 12.1 We] 1251+ 0.41 12.61+ 0.31 13. 9 1+ 1.11 15. 0 1+1. 11 16.31+1.21 2.0 351p<.011 30 4.1 1221 34.3 351211 PA Male 21. cm n1a a1 2131+ 0.31 21 .91+051 231 1+ 1.21 2471+ 17.1 25.11+1.31 1.5 351114.011 Female 4.2 1nl'a1 451+ 0.31 451+ 0.0141+ 0.91 5. 0 1+ 0.5111+ 1. 11 3.5 35 1114.011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period1" Current Overail Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 17 (State REnk} Both 9.41n1a10103101 0.0 01 1281+ 3. 81 11.91.91 12. 6 1+ 0. 71 2.6 ?fa 1p<.051 43 3.2130 W1 34.1 ?fa 123 W1 RI Male 15.41n7a1 1521-0 .21 14.81- 0. 31 21.21+6.41 1912?: 2.01 19..61+041 +2.2 ?fa Female 4.0 1n1a1310.7181+ 0. 41 11+ 1.31001 6.1 1+ 1 .01 3.7 ?fa 1p<.051 Both 12.81n1a1 13.01+ 0.1601+ 1.11 17.71+1.71 2.3 ?fa 1p<.011 23 4.91171 38.3 ?161101 SC Male 21.31n1a1 22.5 1+ 1.21 22.31.11 24. 6 1+ 2. 21 26.11+1.51 28.01+1.91 +1.8 ?fa 1p<.011 Female 5 4 [n7a171-0.7101+ 1.3121+ 0. 21 701+ 0.81 8. 4 1+ 1.41 3.4 ?fa 1p<.051 Both 15.71n1a1 15.81+0. 11 17.1 1+ 1.31 19.31+2.21 1971+ 0.41 22.6 +2.91 +2.5 ?fa 1p<.011 10 +7.01 71 +44.5%1 61 SD Male 27.6 [n1a1 26. 311 .31 27. 9 1+ 1.61 30.1 1+ 2. 21 3201+ 1.91 33. 6 1+ 1.61 1.6 ?fa 1p<.011 Female 4.2 1n1a1 5.81+1.6141+ 0 6131+ 2. 01 31-1 01 11. 31 4. 01 5.8 ?fa 1p<.011 Both 14.61n1a1 1521+ 0.61 16111721+ 0.01 18. 2 1+ 1 .01 +1.4 ?fa 1p<.011 22 3.51281 24.2 ?161311 TN Male 25.1 1n1a1 2541+ 0.31 2681+ 1.31 8 01+ 1. 21 2861+ 0 61 29. 8 1+ 1.21 1.2 ?fa 1p<.011 Female 5.4 [n1a1 631+ 0.91 6 71+ 0 4151+ 0. 81 6.9 (a 0.61 7. 6 1+ 0. 71 1.9 ?fa 1p<.051 Both 12.21nfa1 12.71+ 0.61 12.31- 0.41 13211451+ 0. 91 +1.1 ?fa1p<.011 41 2.31371 18.9 ?161361 TX Male 20.4 [n7a1 2091+ 0.51 20.4 1- 0.61.0123.1 1+ 0. 91 0.9 ?fa 1p<.051 Female 4.8 [his] 5.4 1+ 0.61 5.0 1+ 0.4121+ 0. 21 0.41 6. 4 1+ 0. 81 1.6 ?fa 1p<.051 Both 17.21n1a1 19.0 1+ 1.81 18.21? 0.2521+ 1.21 2.7 ?fa 1p<.011 5 8.01 3 46.5 ?fa 1 4 1W1 UT Male 28.21n1a1 31.1 1+ 2.91 29.4 1- 1.71 32 512.1 ?fa 1p<.051 Female 6.8 [n7a1 7.41+ 0.61 7.51+ 0.2.11 1261+ 2.01 4.4 ?fa 1p<.011 Both 13.21n1a1 16.2 (+3.01 14.91-131 16.61+1.71 18.71+ 2.11 19.71+ 1.01 2.4 ?fa 1p<.011 18 6.41 91 48.6 ?Va 1 21 VT Male 23.6 [n1a1 28. 31+ 61 24.31- 4.01 2731+ 3.01 3101+ 3. 71 3251+ 1.51 1.9 ?fa 1p<.051 Female 4.3 1n1a1 21+4 0. 91 641+ 1.31 6 61+ 0.21 7. 31+ 0. 71 761+ 0.31 3.8 ?fa 1p<.011 Both 12.81n1a1 12.71-011 1291+ 0.31 1361+ 0.71 1461+ 0. 91 1501+ 0.51 +1.2 ?fa 1p<.011 37 2.21391 17.4 ?161411 17.4 Male 21.6 1n1a1 21.31- 1 2101? 0.41 2251+ 1.51 23.61 1 2.1 2391+ 0.21 0.9 ?fa 1p<.051 Female 5.3 [n1a1 5.21? 0.11 591+ 0.71 5.61? 0.31 6. 4 1+ 0 81 691+ 0.51 1.8 ?fa 1p<.051 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (nia) 15411781+ 1.0) 1.1 5511:1405) 24 2.8133) 18.8 c15137) WA Male 24.? [nfa) 2521+ 0 5) 2527.1 1+ 1.1) 13.5 '14 Female 5.0 (We) 8. 4 1+ 0 8) 81+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% 11 +58113) +37.1 i?15114) WV Male 27.2 (nia) 3011+ 2. 9) 5.81 1.+11% Female 5.3 (Na) 551+ 0.581+ 2. 2) 3.7 ?it: Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps01) Female 5.1 (nia0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We.2) 47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (ma) 521+ 0.5) 41+ 5 52) 1+ 1.4) 1251+ 1. 9) 3.2 '14. 1p<.01) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 :Decrease 1.0% I:1lnr:rease 59% - 18.3% - Increase - 29.3% - Increase 31.9% - 37.4% - increase 316% - 57.8% Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age. by State and Sex. Age-Adjusted Annual Rate per 100,000 Persons [Change from Prior Period) Current Overall Overall State Sex Odeled State Rate Change Percent 1999 - MP1: 1 Change 2001 2002 2004 2005 200? 2003 2010 2011 - 2013 2014 - 2013 Rank 1 {Stats Rank} '1 1State Rank) Both 12.3 (rife) 12.? 1+ 0.4) 1291+ 0.2) 1331+ 0.9) 1451+ 0.3) 1541+ 0.9) 1.5 111113401) nr'a 3.1 11113) 25.4 U21) 1nria) LLS. Male 20.9 {rife} 2121+ 0.4) 21.31+ 0.0) 22.51+1.3) 23.51+1.0) 24.51+1.0) 1.1 10111101) Female (11.18] 501+ 0.3) 5.31+ 0.2) 1+ 0.4) 321+ 0.5) 391+ 2.3 '11. 1p<.01) AL Male 25.1 {nits} 2341-1?) 2441+ 1.0) 2341+ 2.0) 291 1+ 1.5) +1.3 ?2310105) Female 5.11n1a) 431+ 0.3) 501+ 0.2) 3.1 1+ 1.1) 341+ 0.3) 2.3 3911:1901) Both 21.0 {013) 2131+ 3.3) 24.21? 0.3) 26.0 1+ 25.41- 0.5] 2331+ 3.4) ?41:34:05) 2 ?.31 4) 3?.4 113) AH Male 33.2 {nl?a} 33.1 1+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 01) 4291+ 2.3) +1.4 91:11:10? Female 3.31n1a) 11.41+ 2.9) 931-13) 11.1 1+ 1.2) 9.9 1- 1.2) 1321+ 3.4) Ms Both 1?.31n1a) 0.7) 19.1 1+ 0.5) 19.1 1- 0.0) 20.41+1.3) 2091+ 0.5) +1.0 5.131090? 15 3.1 132) I13142) AZ Male 29.3 3021+ 1.0) 3031+ 0.4) 30.21? 0.5) 3201+ 1.9) 3241+ 0.4) 0.3 '11] 1p?05) Female 0.4) 321+ 331+ 0.5) 921+ 0.3) 991+ 0.3) 2.2 ?In 1p<.01) Both 15.5 {rife} 1531+ 0.3) 1321+ 0.5) 19.21+1.3) 2121+ 2.0) 2.2 91113101} 12 114) 33.3 ?11: 115) AR Male 23.? 1n1a) 2371+ 0.0) 2121+ 0.5) 2321+ 1.0) 31.? 1+ 3.5) 3351+ 1.9) 1.5 %1ps.05) Female 5.01n1a) 531+ 0.3) 321+ 0.4) 7.5 9151+ 2.1) 3.5 at. 1pc.01) Both 10.3 (11.13] 1131+ 11.01- 0.3) 12.01+1.0) 11.31- 0.1) 12.1 1+ 0.3) 0.9 45 +1.3143) 14.3 ?it: 143) CA Male 1?.91n1a) 1341+ 0.5) 19.1 1+ 1.4) 13.91? 02) 1921+ 0.3) 0.5 110 hits Female 4.1 (nfa) 501+ 0.9) 4.91? 0.1) 5.41+ 0.5) 5.31? 0.1] 531+ 0.3) 101p<.05) Both 1?.31ni?a) 1921+ 1.9) 19.01? 0.2) 20.01+1.0) 21.31+1.5) 23.21+1.3) +1.3 191134.01) 3 5.9112) 34.1 ?it: 122) CID Male 23.3 1n1a) 3091+ 2.3) 30.51- 0.4) 3151+ 1.0) 3341+ 1.9) 3331+ 2.9) +1.4 ?In 1p<.01} Female ?.01n1a) 321+ 1-3) 321+ 0.0) 9.1 1+ 0.9) 10.1 1+ 1.0) 1041+ 0.3) 2.3 39111101) Both 9.31n1a) 3.91? 9.1 1+ 0.2) 1021+ 1.1) 11.01+ 0.3) 1151+ 0.5) 1.3 ?11: 1p?.05) 43 1.9143) 19.2 134) CT Male 13.41n1a) 14.3 1- 1.3) 15.01+ 0.4) 13.31+1.3) 1?.31- 0.3) 0.9 Female 3.31n1a) 331+ 0.2) 0.2) 4.41+ 491+ 0.5) 321+ 1.3) 3.5 ?21.113403 Rates are ageadjusted to the 0.3. year 2000 standard. 1 Model?estimated average annual percentage change based on all reporting periods; p-value indicates statistical signi?cance of trend; his indicates trend not signi?cant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2013. Ranks are from highest rate 11) to lowest rate 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 1 Overall rate change is between the ?rst 11999 2001) and last 12014 2013) reporting periods. Ranks are from largest increase 11) to largest decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is betWeen the ?rst 11999 2001) andlast12014 - 2013) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessanly imply a statistically signi?cant difference. 11 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. an" up- Comment [1111]: Consider sorting state .1 abbreviations alphabetically instead of state Con?uent Insert "National Vital Statistics System" i Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 1090 - 2001 2002 2004 2005 2007 2000 2010 2011 2013 2014 2015 Rank 5 (State Rank} 1' (State Ragnk} Both 15.0 (nfa} 1221+ 1.41 12.01? 0.01 10.0 1+ 1.0} 15.0 1+ 0.0} 10.0 1+ 0.2} 0 0 (prism) 02 0.0 (00} 30.0 111 (50} Male 20.0 {nits} 2001+ 2.151 20.01- 0.4} 23.5 1+ 2.0} 2001+ 0.51 2551+ 0.01 0 0 (111505} Female 5.3 {Na} 0211+ 0211 001+ 0.51 5.0 1+ 0.0} 0.4 1+ 0.5} 021+ 0.51 3.0 (prism) Beth 15.0 (nfa) 15.4 1+ 0.0} 1531+ 0.01 115.3 1+ 0.51 1001+ 12.01 10.0 1+ 0.0} 0.0 (pst?i) 50 0.0 (40} 46.0 ?10 (45} as Male 23.71011} 25.0 1+ 9.3} 3901+ 1101 3501+ 1151 33.1.11 3.51 111.71+ 1.5) 11.9 11, (1111501 1 Female 1100111151111 2.0011000? 531+ 0.01 5.01- 0.3) 0.0 1+ 1.0} 001+ 0.01 3 0 (111505} Both 14.0 {Ma} 15.0 1+ 0.0} 1001+ 0.31 10.3! 1+ 0.01 1021+ 0.01 10.0 1+ 0.1} 0. 0 11:. 1p<. 05 20 5.0 140} 50.0 '10 (40} E11 Male 25.0 {Ma} 20.0 1+ 0.0} 2001+ 0.01 2021+ 201 $131+ 0.01 2001+ 0-01 0. 5 (01} Female 5.0 (nfa} 0.0 1+ 0.5} 0.0 1+ 0.0} 0.1 1+ 0.5} 1.0 1+ 0.0} 7.0 1+ 0.0} 5.0 1p<.01} Both 13.9 {nra} 1301+ 0.31 1301+ 0.91 13.9 1+ 0.9} 10.3 1+ 0.3} 13.0 1+ 2.3) 11.9 11. 1111505) 39 3.3 133) 39.3 11. 130} GA Male 22.0 We} 2201+ 0.01 2101+ 0.01 25.0 1+ 0.0} 20.0 1+ 0.5} 2001+ 5.0} 0.5 n13 Female 5.0 {Na} 4.01- 0.2} 051+ 0151 001+ 0.01 0.5 1+ 0.2} 051+ 0.0} 2.0 1p<. 05} Both 12.01nta} 13.01? 0.0} 1001+ 0.01 15.5 1+ 4.0} 1001+ 0.51 1001+ 0.01 2.0 (111505} 05 0.0 135} 20.0 20 (30} 101E Male 20.0 {nits} 22.0 0.1} 0001+ 1.91 2501+ 5.1} 02.5 1+ 0.5} 2001+ 1.51 2.0 '11. (111505} Female 5.0 {Na} 5.0 0.0} 551+ 0.51 0.0 1+ 0.5} 5.0 1+ 0.01 1101+ 0.31 3.2 (111505 Both 10.0 (nta) 10.0 1+ 0.0} 1001- 0.0} 20.0 1+ 0.3} 20.0 1+ 0.5} 20.0 1+ 0.0} 0.5 '11: 1p<. 05} 40 5125114030} 4110352001410 01} 001} Male 20.0 (nfa} 50.0 1+ 0.1} 331.31- 0.0} 50.0 1+ 0.0} m11+ 0:51 3001+ 031 0.0 (p150 5} Female 13.0 {nits} 0.0 1+ 0.111 0.0 1+ 0.0} 00' 1+ 201 0.5 1+ 0.5} 1401+ 0.0} 4.0 '11. 1p<. 05} Both 0.0 {hrs} 001+ 0.121 0.4? 1+ 0.01 10.0 1+ 0.0} 10.0 1+ 0.5} 10.0 1+ 0.0} 2.5 11:. 1p<. 05} 40 ++220211100111 03023810112020} MIA Male 12.1 {Ma} 1001+ 0.51 1021+ 0.51 15.5 1+ 2.4} 1051+ 0.01 10.0 1+ 0.0} 2.0 1p<. 05} Female 3.0 (nfa} 201 0.0} 0.0 1+ 0.0} 0211+ 0.111 4.0 1+ 0.0} 001+ 0.51 3.0 1p<. 05} Both 13.9103} 13.5 1+ 0.1} 13.9 1+ 0.1} 1391+ 0.0} 1001+ as) 13.0 1+ 11.1} 1.9 11. 1p<. 01} 33 3.9126) 33.9 11.129} itll Male 20.0 We} 20.0 1+ 0.0} 24.5 1+ 0.2} 20.0 1+ 0.0} 20.0 1+ 2.0} 20.0 1+ 1.0} 1.5 1p<. 01} Female 4.5 {Na} 5.0 1+ 0.4} 5.0 1+ 0.2} 5.0 1+ 0.0} 0.0 1+ 0.0} 0151+ 0.01 2.0 1p<. 01} Rates are age-adjusted to the U.S. year 2000 standard. 1 Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not signi?cant. Current state rank 150 states and the District of Columbia} is for the reporting period 2014 .. 2010. Ranks are from highest rate to lowest rate 1511. ranks do not necessarily imply a statistically signi?cant difference. 1' Overall rate change is between the first (1000 2001} and last (2014 2016} reporting periods. Ranks are from largest increase to largest decrease Differences between ranks do not necessarily imply a statistically signi?cant difference. Overall percent change in rates is between the ?rst (1000 - 2001 and last (2014 2016} reporting periods. Ranks are from largest percentage increase to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 0 Rate based on 20 suicides. Differences between Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last pen'ods and might have contributed to lower reported rates. 1? Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 11.-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 (We) 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 41n1a1 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 (Ma) 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.51n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20151 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51n7a) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30 Till) 34.1 ?fa (23 RI Male 15.4 (nfa) .2) 14.8(- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.16.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (nfa) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (nfa) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18.2 1.0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 26.8 1.3) 8 0 1. 2) 28.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (nfa) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (nfa) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3 1W) 46.5 3?a( 4 UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (nfa) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2(+4 0. 9) 6.4 1.3) 6 6 0.2) 7. 3 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (nfa) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 25.1 1+1 .0) 25. a 1+ 1:1. 9) 27.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nl?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 Decrease 1.0% I:I ncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 37.8% - 57.8% Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age. by State and Sex. Age-Adjusted Annual Rate per 100,000 Persons [Change from Prior Period) Current Overall Overall State Sex Odeled State Rate Change Percent 1999 - MP1: 1 Change 2001 2002 2004 2005 200? 2003 2010 2011 - 2013 2014 - 2013 Rank 1 {Stats Rank} '1 1State Rank) Both 12.3 (rife) 12.? 1+ 0.4) 1291+ 0.2) 1331+ 0.9) 1451+ 0.3) 1541+ 0.9) 1.5 111113401) nr'a 3.1 11113) 25.4 U21) 1nria) LLS. Male 20.9 {rife} 2121+ 0.4) 21.31+ 0.0) 22.51+1.3) 23.51+1.0) 24.51+1.0) 1.1 10111101) Female (11.18] 501+ 0.3) 5.31+ 0.2) 1+ 0.4) 321+ 0.5) 391+ 2.3 '11. 1p<.01) AL Male 25.1 {nits} 2341-1?) 2441+ 1.0) 2341+ 2.0) 291 1+ 1.5) +1.3 ?2310105) Female 5.11n1a) 431+ 0.3) 501+ 0.2) 3.1 1+ 1.1) 341+ 0.3) 2.3 3911:1901) Both 21.0 {013) 2131+ 3.3) 24.21? 0.3) 26.0 1+ 25.41- 0.5] 2331+ 3.4) ?41:34:05) 2 ?.31 4) 3?.4 113) AH Male 33.2 {nl?a} 33.1 1+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 01) 4291+ 2.3) +1.4 91:11:10? Female 3.31n1a) 11.41+ 2.9) 931-13) 11.1 1+ 1.2) 9.9 1- 1.2) 1321+ 3.4) Ms Both 1?.31n1a) 0.7) 19.1 1+ 0.5) 19.1 1- 0.0) 20.41+1.3) 2091+ 0.5) +1.0 5.131090? 15 3.1 132) I13142) AZ Male 29.3 3021+ 1.0) 3031+ 0.4) 30.21? 0.5) 3201+ 1.9) 3241+ 0.4) 0.3 '11] 1p?05) Female 0.4) 321+ 331+ 0.5) 921+ 0.3) 991+ 0.3) 2.2 ?In 1p<.01) Both 15.5 {rife} 1531+ 0.3) 1321+ 0.5) 19.21+1.3) 2121+ 2.0) 2.2 91113101} 12 114) 33.3 ?11: 115) AR Male 23.? 1n1a) 2371+ 0.0) 2121+ 0.5) 2321+ 1.0) 31.? 1+ 3.5) 3351+ 1.9) 1.5 %1ps.05) Female 5.01n1a) 531+ 0.3) 321+ 0.4) 7.5 9151+ 2.1) 3.5 at. 1pc.01) Both 10.3 (11.13] 1131+ 11.01- 0.3) 12.01+1.0) 11.31- 0.1) 12.1 1+ 0.3) 0.9 45 +1.3143) 14.3 ?it: 143) CA Male 1?.91n1a) 1341+ 0.5) 19.1 1+ 1.4) 13.91? 02) 1921+ 0.3) 0.5 110 hits Female 4.1 (nfa) 501+ 0.9) 4.91? 0.1) 5.41+ 0.5) 5.31? 0.1] 531+ 0.3) 101p<.05) Both 1?.31ni?a) 1921+ 1.9) 19.01? 0.2) 20.01+1.0) 21.31+1.5) 23.21+1.3) +1.3 191134.01) 3 5.9112) 34.1 ?it: 122) CID Male 23.3 1n1a) 3091+ 2.3) 30.51- 0.4) 3151+ 1.0) 3341+ 1.9) 3331+ 2.9) +1.4 ?In 1p<.01} Female ?.01n1a) 321+ 1-3) 321+ 0.0) 9.1 1+ 0.9) 10.1 1+ 1.0) 1041+ 0.3) 2.3 39111101) Both 9.31n1a) 3.91? 9.1 1+ 0.2) 1021+ 1.1) 11.01+ 0.3) 1151+ 0.5) 1.3 ?11: 1p?.05) 43 1.9143) 19.2 134) CT Male 13.41n1a) 14.3 1- 1.3) 15.01+ 0.4) 13.31+1.3) 1?.31- 0.3) 0.9 Female 3.31n1a) 331+ 0.2) 0.2) 4.41+ 491+ 0.5) 321+ 1.3) 3.5 ?21.113403 Rates are ageadjusted to the 0.3. year 2000 standard. 1 Model?estimated average annual percentage change based on all reporting periods; p-value indicates statistical signi?cance of trend; his indicates trend not signi?cant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2013. Ranks are from highest rate 11) to lowest rate 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 1 Overall rate change is between the ?rst 11999 2001) and last 12014 2013) reporting periods. Ranks are from largest increase 11) to largest decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is betWeen the ?rst 11999 2001) andlast12014 - 2013) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessanly imply a statistically signi?cant difference. 11 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. an" up- Comment [1111]: Consider sorting state .1 abbreviations alphabetically instead of state Con?uent Insert "National Vital Statistics System" i Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 1090 - 2001 2002 2004 2005 2007 2000 2010 2011 2013 2014 2015 Rank 5 (State Rank} 1' (State Ragnk} Both 15.0 (nfa} 1221+ 1.41 12.01? 0.01 10.0 1+ 1.0} 15.0 1+ 0.0} 10.0 1+ 0.2} 0 0 (prism) 02 0.0 (00} 30.0 111 (50} Male 20.0 {nits} 2001+ 2.151 20.01- 0.4} 23.5 1+ 2.0} 2001+ 0.51 2551+ 0.01 0 0 (111505} Female 5.3 {Na} 0211+ 0211 001+ 0.51 5.0 1+ 0.0} 0.4 1+ 0.5} 021+ 0.51 3.0 (prism) Beth 15.0 (nfa) 15.4 1+ 0.0} 1531+ 0.01 115.3 1+ 0.51 1001+ 12.01 10.0 1+ 0.0} 0.0 (pst?i) 50 0.0 (40} 46.0 ?10 (45} as Male 23.71011} 25.0 1+ 9.3} 3901+ 1101 3501+ 1151 33.1.11 3.51 111.71+ 1.5) 11.9 11, (1111501 1 Female 1100111151111 2.0011000? 531+ 0.01 5.01- 0.3) 0.0 1+ 1.0} 001+ 0.01 3 0 (111505} Both 14.0 {Ma} 15.0 1+ 0.0} 1001+ 0.31 10.3! 1+ 0.01 1021+ 0.01 10.0 1+ 0.1} 0. 0 11:. 1p<. 05 20 5.0 140} 50.0 '10 (40} E11 Male 25.0 {Ma} 20.0 1+ 0.0} 2001+ 0.01 2021+ 201 $131+ 0.01 2001+ 0-01 0. 5 (01} Female 5.0 (nfa} 0.0 1+ 0.5} 0.0 1+ 0.0} 0.1 1+ 0.5} 1.0 1+ 0.0} 7.0 1+ 0.0} 5.0 1p<.01} Both 13.9 {nra} 1301+ 0.31 1301+ 0.91 13.9 1+ 0.9} 10.3 1+ 0.3} 13.0 1+ 2.3) 11.9 11. 1111505) 39 3.3 133) 39.3 11. 130} GA Male 22.0 We} 2201+ 0.01 2101+ 0.01 25.0 1+ 0.0} 20.0 1+ 0.5} 2001+ 5.0} 0.5 n13 Female 5.0 {Na} 4.01- 0.2} 051+ 0151 001+ 0.01 0.5 1+ 0.2} 051+ 0.0} 2.0 1p<. 05} Both 12.01nta} 13.01? 0.0} 1001+ 0.01 15.5 1+ 4.0} 1001+ 0.51 1001+ 0.01 2.0 (111505} 05 0.0 135} 20.0 20 (30} 101E Male 20.0 {nits} 22.0 0.1} 0001+ 1.91 2501+ 5.1} 02.5 1+ 0.5} 2001+ 1.51 2.0 '11. (111505} Female 5.0 {Na} 5.0 0.0} 551+ 0.51 0.0 1+ 0.5} 5.0 1+ 0.01 1101+ 0.31 3.2 (111505 Both 10.0 (nta) 10.0 1+ 0.0} 1001- 0.0} 20.0 1+ 0.3} 20.0 1+ 0.5} 20.0 1+ 0.0} 0.5 '11: 1p<. 05} 40 5125114030} 4110352001410 01} 001} Male 20.0 (nfa} 50.0 1+ 0.1} 331.31- 0.0} 50.0 1+ 0.0} m11+ 0:51 3001+ 031 0.0 (p150 5} Female 13.0 {nits} 0.0 1+ 0.111 0.0 1+ 0.0} 00' 1+ 201 0.5 1+ 0.5} 1401+ 0.0} 4.0 '11. 1p<. 05} Both 0.0 {hrs} 001+ 0.121 0.4? 1+ 0.01 10.0 1+ 0.0} 10.0 1+ 0.5} 10.0 1+ 0.0} 2.5 11:. 1p<. 05} 40 ++220211100111 03023810112020} MIA Male 12.1 {Ma} 1001+ 0.51 1021+ 0.51 15.5 1+ 2.4} 1051+ 0.01 10.0 1+ 0.0} 2.0 1p<. 05} Female 3.0 (nfa} 201 0.0} 0.0 1+ 0.0} 0211+ 0.111 4.0 1+ 0.0} 001+ 0.51 3.0 1p<. 05} Both 13.9103} 13.5 1+ 0.1} 13.9 1+ 0.1} 1391+ 0.0} 1001+ as) 13.0 1+ 11.1} 1.9 11. 1p<. 01} 33 3.9126) 33.9 11.129} itll Male 20.0 We} 20.0 1+ 0.0} 24.5 1+ 0.2} 20.0 1+ 0.0} 20.0 1+ 2.0} 20.0 1+ 1.0} 1.5 1p<. 01} Female 4.5 {Na} 5.0 1+ 0.4} 5.0 1+ 0.2} 5.0 1+ 0.0} 0.0 1+ 0.0} 0151+ 0.01 2.0 1p<. 01} Rates are age-adjusted to the U.S. year 2000 standard. 1 Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not signi?cant. Current state rank 150 states and the District of Columbia} is for the reporting period 2014 .. 2010. Ranks are from highest rate to lowest rate 1511. ranks do not necessarily imply a statistically signi?cant difference. 1' Overall rate change is between the first (1000 2001} and last (2014 2016} reporting periods. Ranks are from largest increase to largest decrease Differences between ranks do not necessarily imply a statistically signi?cant difference. Overall percent change in rates is between the ?rst (1000 - 2001 and last (2014 2016} reporting periods. Ranks are from largest percentage increase to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 0 Rate based on 20 suicides. Differences between Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last pen'ods and might have contributed to lower reported rates. 1? Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 11.-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 (We) 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 41n1a1 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 (Ma) 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.51n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20151 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51n7a) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30 Till) 34.1 ?fa (23 RI Male 15.4 (nfa) .2) 14.8(- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.16.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (nfa) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (nfa) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18.2 1.0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 26.8 1.3) 8 0 1. 2) 28.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (nfa) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (nfa) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3 1W) 46.5 3?a( 4 UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (nfa) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2(+4 0. 9) 6.4 1.3) 6 6 0.2) 7. 3 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (nfa) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 25.1 1+1 .0) 25. a 1+ 1:1. 9) 27.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nl?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 Decrease 1.0% I:I ncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 37.8% - 57.8% Table 1. Trends in Suicide Rates among Persons :2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2991 2992 2994 2995 2997 2998 2919 2911 2913 2914 2916 Rank (State Rank) '9 (State Ragnk) H. Both 12.3 (We) 12.7 9.4) 12.9 9.2) 13.8 9.9) 14.5 9.8) 15.4 9.9) 1.5 ?70 (pr-1.91) nia 3.1 (nia) 25.4 (his) U.S. Male 29.9 (nia) 21.2 9.4) 21.3 9.9) 22.5 23.5 1.9) 24.5 1.9) Female 4.7 5.9 9.3) 5.3 9.2) 5.7 9.4) 6.2 9.5) 6.9 9.7) 2.6 31. Both 14.3 (nia) 13.4 (- 9.9) 14.1 9.6) 15.6 1.6) 16.4 9.7) 17.5 .1) +1.6 (ps. 95) 25 3.1 (31) 21.9 (33) AL Male 25.1 (nia) 23.4 (- 1.7) 24.4 1.9) 26.4 2.9) 27.6 1.1) 29.1 1. 5) 1. 3 (ps. 95) Female 5.1 (nia) 4.8 (- 9.3) 5.9 9.2) 1 6.4 9.3) 9. 7) 2. 6 (.ps 91) Both 21.9 (nia) 24.8 3.8) 24.2 (- 9.6) 26.9 1.7) 25.4 (- 9.5) 28.8 3.4) +1 7 (p4. 95) 2 7.8 4) 37.4 (13) AK Male 33.2 (nia) 38.1 4.9) 38.9 9.8) 49.1 49.1 (- 9.1) 42. 9 2. 8) +1.4 (ps. 91) Female 8.6 (nia) 11.4 2.9) 9.8 (- 1.6) 11.1 9.9 (- 1.his Both 17.8 (nia) 18.5 9.7) 19.1 9.5) 19.1(- 9.9) 29.4 1.3) 29.9 9. 5) +1.9 91) 15 3.1 (32) 17.3 (42) .42 Male 29.3 (nia) 39.2 1.9) 39. 6 9.4) 39. 2 9.5) 32. 9 1.(ps. 95) Female 7.1 (nia) 7.5 9.4) 9.7) 9.5) 9.6) 9. 6) 2. 2 (ps. 91) Both 15.5 (nia) 15.8 9.3) 16.2 9.5) 17.6(+ 1.4) 19.2 1.6) 21 91) 12 +5.7(14) AR Male 26 7 (nia) 26.7 9.9) 27. 2 9.5) 28. 2 1.9) 31. 7 3.5) 33. 5 1. 9) 1.6 (ps. 95) Female 5.6 (nia) 5.9 9.3) 6. 2 9 4) 1.7) 7.5 (- 9.4) 2.1) 3. 6 (.ps 91) Both 19.6 (nia) 11.3 11 93) 11.8 (- 9.1) 12.1 95) 45 1.6 (46) CA Male 17.9 (nia) 18.4 9.5) 17.7-( 9.7) 19.1 18.9 (-9.2) +95% his Female 4.1 (nia) 5.9 9.9) 4.9 9.1) 9.5) 5. 3 (- 9.1) 9. 3) 7 (ps. 95) Both 17.3 (nia) 19. 9 .2) 29. 9 1.9) 21.6 1.5) 23. 2 16) 1.8 (p4. 91) 8 5.9 (12) 34.1 (22) CO Male 28.6 (nia) 39.9 2.3) 39.5 (9 .4) 31.5 1.9) 33.4 1.9) 36. 3 2. 9) +1.4 (ps. 91) Female 7.9 (nia) 8.2 1.3) 8.2 9.(ps. 91) Both 9.6 (nia) 8.9 (w 9.7) 9.1 9.2) 19. 2 1.1) 11.9 9.8) 11.5 9.5) +1.6 (cs. 95) 46 1.9 (43) +192 (34) CT Male 16.4 (nia) 14.6 (w 1.8) 15.9 9.4) 16. 6 1.6) 17. 6 1.9) 17.3 (- 9.3) 9.9 his Female 3.6 (nia) 3.8 9.2) 3.7 (- 9.2) 9.7) 9.5) 6.2 1.3) 3.5 Rates are age-adjusted to the US. year 2999 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase to largest percentage decrease (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 29 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 1090 - 2001 2002 2004 2005 2007 2000 2010 2011 2013 2014 2015 Rank 5 (State Rank} 1' (State Ragnk} Both 15.0 (nfa} 1221+ 1.41 12.01? 0.01 10.0 1+ 1.0} 15.0 1+ 0.0} 10.0 1+ 0.2} 0 0 (prism) 02 0.0 (00} 30.0 111 (50} Male 20.0 {nits} 2001+ 2.151 20.01- 0.4} 23.5 1+ 2.0} 2001+ 0.51 2551+ 0.01 0 0 (111505} Female 5.3 {Na} 0211+ 0211 001+ 0.51 5.0 1+ 0.0} 0.4 1+ 0.5} 021+ 0.51 3.0 (prism) Beth 15.0 (nfa) 15.4 1+ 0.0} 1531+ 0.01 115.3 1+ 0.51 1001+ 12.01 10.0 1+ 0.0} 0.0 (pst?i) 50 0.0 (40} 46.0 ?10 (45} as Male 23.71011} 25.0 1+ 9.3} 3901+ 1101 3501+ 1151 33.1.11 3.51 111.71+ 1.5) 11.9 11, (1111501 1 Female 1100111151111 2.0011000? 531+ 0.01 5.01- 0.3) 0.0 1+ 1.0} 001+ 0.01 3 0 (111505} Both 14.0 {Ma} 15.0 1+ 0.0} 1001+ 0.31 10.3! 1+ 0.01 1021+ 0.01 10.0 1+ 0.1} 0. 0 11:. 1p<. 05 20 5.0 140} 50.0 '10 (40} E11 Male 25.0 {Ma} 20.0 1+ 0.0} 2001+ 0.01 2021+ 201 $131+ 0.01 2001+ 0-01 0. 5 (01} Female 5.0 (nfa} 0.0 1+ 0.5} 0.0 1+ 0.0} 0.1 1+ 0.5} 1.0 1+ 0.0} 7.0 1+ 0.0} 5.0 1p<.01} Both 13.9 {nra} 1301+ 0.31 1301+ 0.91 13.9 1+ 0.9} 10.3 1+ 0.3} 13.0 1+ 2.3) 11.9 11. 1111505) 39 3.3 133) 39.3 11. 130} GA Male 22.0 We} 2201+ 0.01 2101+ 0.01 25.0 1+ 0.0} 20.0 1+ 0.5} 2001+ 5.0} 0.5 n13 Female 5.0 {Na} 4.01- 0.2} 051+ 0151 001+ 0.01 0.5 1+ 0.2} 051+ 0.0} 2.0 1p<. 05} Both 12.01nta} 13.01? 0.0} 1001+ 0.01 15.5 1+ 4.0} 1001+ 0.51 1001+ 0.01 2.0 (111505} 05 0.0 135} 20.0 20 (30} 101E Male 20.0 {nits} 22.0 0.1} 0001+ 1.91 2501+ 5.1} 02.5 1+ 0.5} 2001+ 1.51 2.0 '11. (111505} Female 5.0 {Na} 5.0 0.0} 551+ 0.51 0.0 1+ 0.5} 5.0 1+ 0.01 1101+ 0.31 3.2 (111505 Both 10.0 (nta) 10.0 1+ 0.0} 1001- 0.0} 20.0 1+ 0.3} 20.0 1+ 0.5} 20.0 1+ 0.0} 0.5 '11: 1p<. 05} 40 5125114030} 4110352001410 01} 001} Male 20.0 (nfa} 50.0 1+ 0.1} 331.31- 0.0} 50.0 1+ 0.0} m11+ 0:51 3001+ 031 0.0 (p150 5} Female 13.0 {nits} 0.0 1+ 0.111 0.0 1+ 0.0} 00' 1+ 201 0.5 1+ 0.5} 1401+ 0.0} 4.0 '11. 1p<. 05} Both 0.0 {hrs} 001+ 0.121 0.4? 1+ 0.01 10.0 1+ 0.0} 10.0 1+ 0.5} 10.0 1+ 0.0} 2.5 11:. 1p<. 05} 40 ++220211100111 03023810112020} MIA Male 12.1 {Ma} 1001+ 0.51 1021+ 0.51 15.5 1+ 2.4} 1051+ 0.01 10.0 1+ 0.0} 2.0 1p<. 05} Female 3.0 (nfa} 201 0.0} 0.0 1+ 0.0} 0211+ 0.111 4.0 1+ 0.0} 001+ 0.51 3.0 1p<. 05} Both 13.9103} 13.5 1+ 0.1} 13.9 1+ 0.1} 1391+ 0.0} 1001+ as) 13.0 1+ 11.1} 1.9 11. 1p<. 01} 33 3.9126) 33.9 11.129} itll Male 20.0 We} 20.0 1+ 0.0} 24.5 1+ 0.2} 20.0 1+ 0.0} 20.0 1+ 2.0} 20.0 1+ 1.0} 1.5 1p<. 01} Female 4.5 {Na} 5.0 1+ 0.4} 5.0 1+ 0.2} 5.0 1+ 0.0} 0.0 1+ 0.0} 0151+ 0.01 2.0 1p<. 01} Rates are age-adjusted to the U.S. year 2000 standard. 1 Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not signi?cant. Current state rank 150 states and the District of Columbia} is for the reporting period 2014 .. 2010. Ranks are from highest rate to lowest rate 1511. ranks do not necessarily imply a statistically signi?cant difference. 1' Overall rate change is between the first (1000 2001} and last (2014 2016} reporting periods. Ranks are from largest increase to largest decrease Differences between ranks do not necessarily imply a statistically signi?cant difference. Overall percent change in rates is between the ?rst (1000 - 2001 and last (2014 2016} reporting periods. Ranks are from largest percentage increase to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. 0 Rate based on 20 suicides. Differences between Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last pen'ods and might have contributed to lower reported rates. 1? Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 11.-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 (We) 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 41n1a1 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 (Ma) 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.51n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20151 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51n7a) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30 Till) 34.1 ?fa (23 RI Male 15.4 (nfa) .2) 14.8(- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.16.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (nfa) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (nfa) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18.2 1.0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 26.8 1.3) 8 0 1. 2) 28.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (nfa) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (nfa) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3 1W) 46.5 3?a( 4 UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (nfa) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2(+4 0. 9) 6.4 1.3) 6 6 0.2) 7. 3 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (nfa) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 25.1 1+1 .0) 25. a 1+ 1:1. 9) 27.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nl?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 Decrease 1.0% I:I ncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 37.8% - 57.8% OJ woo-4Rev 4.5.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedii occurred in the United States among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10"h leading cause of death and is one ofjust three leading causes that are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The National Strotegyfor Suicide Prevention{NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts are largely clinically-oriented, focused on treating mental health problems and preventing re?attempts (6). Apart from MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor coping and problemrsolving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes Rev 4.5.18, e-clearance state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi?Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Classification of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS [665] The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicideesuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by Fiev 4.5.13, e-clearance R4 [26.9% vs 31.3%) or poisoning [10.4% vs 19.8%) than those with known I. Comment i would add in the ?li: 85 These differences remained significant in the adjusted models. it for opioids as it is timely and relevant 36 Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology if commit!? lthGUEht they were 87? results, decedents without known l'vll-lP were less likely to test positive for any substance overall 95% like?? to be from other I facility which Included drug rehab? E48 but more likely to test positive for alcohol [aGFi=1.2. 95% [59 All suicide decedents with lleP and approximately 85% without ,I'r 9i} circumstances information [Table People without known MHP were less likely to have any substance abuse 5 9 problems 95% While two-thirds of those with known MHP had a history of mental health Ii 92 or substance abuse treatment just over half were in current treatment. 5 93 Decedents without, versus those with, known MHP, had significantly greater likelihood of any relationship 94 problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. ,5 95 [12.5% and recently perpetratlng interpersonal violence vs. They also were more likely 96 to have experienced any life stressors [54.2% vs such as criminal~legal problems [10.2% vs. or ll 9? evictionfloss of home vs. and were more likely to have had a crisis within the preceding or 98 upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. 99 Among all people with crises, intimate partner problems were the most common types and did not differ by IUD group. Similarly, among people without versus with MHP, physical health problems {23.2% and 21.4%] and I i i IOI job/financial problems [15.5% and 16.3%) were commonly experienced and did not differ by group. fi2 Decedents without known MHP had signi?cantly lower odds of recent release from any institution MB 95% out among those who were recently released they were significantly more likely to be Iii-4 released from a correctional facility [25.2% vs. or hospital vs. 33.0%] than those with a known [on group were released from facilities. Ill".F Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of [lift suicidal ideation [23.0% vs. 40.3%] and prior suicide attempts vs. More than one in five people in both groups disclosed suicide intent [22.4% vs. I Ill Conclusions and Comments From 1999-2015, 44 states saw significant increases in suicide rates and 25 states experienced substantial l2 increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and l3 females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- I [4 female suicide rates between 1999-2014 Additional research into the specific causes ofthese trends is [5 necessary. Fortunately, data from the 22 states participating in provides important insight into suicide no circumstances and can help states identify prevention priorities. ll? Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention research and practice is heavily oriented towards treatment of lv'lHP and prevention of reattempts. ?9 Additional focus on non?mental health factors, further upstream, is essential to a public health approach as l2ll the current study found that more than half of suicide decedents in did not have a knowh MHP. This l2l group suffered more from relationship problems and other life stressors such as criminal?legal matters, 122 evictionfloss of home, and recent or impending crises. l23 Similarly, people with lleP also experienced other life stressors such relationship, andfor 24 physical health problems. These findings point to the need to both prevent the Conditions associated with 3 125 126 127 128 129 131} 131 132 133 134 135 .136 137 138 139 140 141 142 143 145 146 147V 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 Rev 4.5.13, e-clearance mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two-thirds of this group had a history of any mental health and/or substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor?patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physical/mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward), or decreased percentages of such deaths overtime (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included rapresent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Paiicies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. 4 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 1% 197 198 199 200 201 202 203 204 205 206 207 Rev 4.5.13, e-clearance Corresponding author: Deborah M. Stone, 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atia nta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca Ml. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy 5, Xu 1, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wesserman D, van Heeringen K, Arensman E, Sarchia pone M, et 31. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 201?. 3. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed 1, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 2005;361511491-510. Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.5.18 (Tables 2&3} Word Count: 186871800 Rev 3.23.18, e-clearance Comment IFCH: I think some sort of clarification is needed to distinguish from previous two sentences since what is reported here are not data on full examined time period or all 50 states. Tracked an idea but there are likely other ways to clarify. ,[Com ment In?ll: Updated 3 1 I Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates ?1 2 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott Fl. Kegler, Keming Yuan, 3 Kristin M. Holland, PhD,"l Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 4 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mentalheal?thpsoblemsli?HPi 5 state-level trends in,saieide and eEheethe multiple 6 contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six '8 consecutive three-year periods (1999-2016}, were assessed using data from the National Vital Statistics System I 9 for 50 states and Washington, D.C. Data from the National Violent Death Reporting System covering 10 27 states in 2015, were used to examine contributing circumstances among decedents with and without known - ll MHPmental health problems l; [2 Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. 5 l4 Home; decedents in 27 states, 54.0% not have a known IS with circumstance information, several circumstances, including relationship I6 problemsfloss [45.1% vs life stressors [54.2% vs 491%], health problems l23.2 vs 21.4%} and IT recentfimpending crises {32.9% vs were significantly more likely among those without a known l'leP 18 than decedents with MHP, but were common across groups. l9 Conclusions: Suicide rates increased significantly across most states from 1999-2015. Various circumstances 20 contributed to suicides among people with and without known MHP. 2 Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach 22 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 23 after a suicide occurs. 24- INTRODUCTION 25 BACKGROUND AND PURPOSE 26 In 2016, nearly 45,000 suicides {15.6i100,000 [age-adjusted? occurred in the United States among people I 210 years old ill. Between 1999 and 2015, suicide rates increased across ?sexes, raciallethnic groups, and 28 urbanization levels (2, Suicide is the 10th leading cause of death and is one of just three leading causes that 29 are increasing Additionally. rates of Emergency Department visits for nonfatal self-harm, a key risk factor 30 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 3 nation more than $69 billion in direct medical and work loss costs ill. 32 The National Strotegyfor Suicide calls for a public health approach to suicide prevention 33 with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 34 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 35 Despite the NSSP guidance, suicide prevention efforts largely focus on 36 interventionsmental?health focused on mental health problems oe?eaeviding 3? Letsttempesaiede l?l- seamen! these roles gather 38 contributing circumstances include social isolation, relationship problemspancl economic problems, access to 39 lethal means substances, firearms, bridges) among people at risk, and poor coping and problem?solving 40 Expanded awareness of thes_e additional circumstances that?contributi?ge 1 Fiev 3.23.18, e-clearance 4 to suicide risk apart?feam?Mluliirand action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-speci?c trends in suicide 42 43 rates, assesses the multiple contributing factors to guicide, and provides recommendations for multi-level 44 comprehensive suicide prevention. 45 METHODS 46 Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in younger children 4? are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death 48 certificate records {international ClosSificotion of Diseases 10?? Revision, underlying-cause-of death codes K60- 49 X84, 003}. Age-specific population estimates were obtained from US. Census Bureaulli-lational Center for 50 Health Statistics bridged-race population data releases. SI 52 National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods 53 spanning 1999?2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed 54 per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?year data 55 aggregates, employing weighted least squares regression with inverse?variance weighting. Modeled rate trends 56 are reported in terms of average annual percentage changes 5? 53 Characteristics?nd circumstances of suicide decedents 210 years, with and without known I'leP, were '1 Comment Suggest pulling out the references to tables 2 and 3 since table 1 hasn't been mentioned yet and helps saves on word count. lfyou want to keep them in, suggest adding ?table 1? somewhere into the above paragraph. Comment Iaal?JI: Took the ref to tables 1-i- Comment Given attention to this 5 particular at~risk population in the results, made me wonder if some mention of needs of this group should be in the ,i discussion. There are other higher risk 5 groups described below. A consideration i could be in the discussion about I prevention needs of underserved areas {bottom of page 3} some phrasing could he added about the need for services for 5 higher risk groups, such as those who have served in the military and those who have difficulties, such as MPH, intimate partner problems, have physical health 59 compared in the 27 states with complete data participating in National Violent Death Reporting System 60 in 2015. MUD-RS defines MHP as disorders and listed in the Diagnostic and Statistical Manual 6] of Mental Disorders, Fifth Edition with the exception ofalcohol and other substance dependence, which are 62 captured separately in NUDRS. aggregates data from three primary data sources: death certificates, 63 coroner/medical examiner reports {including toxicology}, and law enforcement reports. Decedents with and 64 without known were compared using Chi?square tests. Logistic regression analyses estimated adjusted 65 odds ratios with 95% confidence intervals controlling for age group, sex, and race/ethnicity. 66 RESULTS 67 The most recent overall suicide rates {representing 2014?2015} varied four?fold, from 5.9 to 29.2 68 {Montana} per 100,000 persons per year {Table Across the study period, rates increased in all states, except 69 Nevada {which had a consistently high rate throughout}, with absolute increases ranging from +0.3 {Delaware} to +8.1 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% {Delaware} to {North ?i?l Dakota}, with increases of more than 30% observed in 25 states {Table 1, Figure T2 ?3 Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females {43 ?4 states}, as well as for the US. overall {Table Nationally, the model-estimated AAPC forthe overall suicide T5 rate was By sex, estimated national rate trends further indicated significant increases for males 76 and females problems, and have ?nancial and legal dif?culties. Or on page 4 where "identifying and better supporting people Lat risk? is stated an e.g. could be added. Comment [21:19}: Did not end up adding this as it seemed odd to call out this one Lpopulation. 7?7 7'8 Suicide decedents without known MHP were compared to those with MHP in 2? states. T9 While all decedents were predominately male {Table 2; 76.8%} and non-Hispanic white those without 80 known MHP, relative to those with MHP, were more likely male {83.6% vs. 63.8%; odds ratio 95% CI 2 8 2.2?2.5} and racialfethnic minorities range: Suicide decedents without known MHP also had 82 significantly greater odds of perpetrating homicide?suicide {adjusted odds ratio 2.9, 95% CI 83 Among adult decedents, 20.1% and 15.3% of people without and with respectively, ever served, or were 34 currently Serving, in the 0.5. milita 199 192 I93 194 l05 lil? lll24 125 Rev 3.23.18, e-clearance while firearms were the most common method of suicide overall and for both groups! decedents Comment Well: It appears without known l'leP were more likely to die by firearm (55.3% vs. 40.6%} and less likely to die by x" being compared to non-MHP. Tracked {25.9% vs 31.3%] or poisoning [10.4% vs 19.3%} than those with known what i think the Edit i5 bl-lt PIEEISE dOUblE These differences remained significant in the adjusted models. ?my Chad? . . Comment [21119]: You are correct. Should Decedents Without known MHP were less likely to receive toxicology testing. Among those with toxicology be with in the second instances not results, decedents without known MHP were less likely to test positive for any substance overall 95% without. but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table People without known MHP were less likely to have any substance abuse problems 95% or io?of?nanclal problems 95% Cl While two-thirds of those with known lleP had a history of mental health or substance abuse treatment over half were in current treatment. Decedents without known MHP versus those with known MHP had significantly great-9r likelihood of any relationship problemfloss [45.1% vs. speci?cally intimate partner problems [30.2% vs. arguments/conflicts [17.5% vs. and recently perpetrating interpersonal violence vs. They vvere also more likely to have experienced any life stressors [54.2% vs 493%}, such as criminal-legal problems (10.7% vs. or evictionfloss of home vs. or health problems i23.2% and 21.4%} and they were more likely to have had a crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. Decedents without known MHP had signi?cantly lower odds of recent release from any institution laClR=D.5. 95% but among those who were recently released they were significantly more likely to be released from a correctional facility vs. or hospital (43.7% vs. 33.0%} than those with a known MHP. Among decedents with known MHP who were recently released from an institution of this group were released from facilities. Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%} and prior suicide attempts {10.3% vs. More than 93% in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999?2015, 44 states saw significant in suicide rates and 25 states experienced rate increases of more than 30%. Rates increased significantly for?mgmales, {in 34 states}; and far?females;1in 43 states}. This ?nding is consistent with prior research showing a decreasing gender gap in male?female suicide rates between research into the specific caUses of these trends is necessary. Fortunately, data from the 2? states participating in important insight into the suicide circumstances and help slat?es identify prevention regularly state that fsuicide _'is not caused by a single factor;' however. suiide prevention research and practice-prevention weighte_d towards activities focused on MHP la}. in contrast, tIhe current study found that more than half of suicide decedents in did not have a known MHP. This group 3 Rev 3.28.18, e-clearance 126 experienced soff-er-ed?meserelationship problems-i_ssues and o_ther life stressors such as criminal legal-reassess 127 and health problems. They also experienced moreand recentg?eeimpending crises I28 Lwhich may indicate suicides took place I29 minimal deliberation (l?iL? Similarly: [30 Among?people with MHF struggled more with iobi?financial problems and substance abuse and also commonly [31 experienced intimate partner, health problems. and other life stressors. These findings all I32 align with a large literature suggesting vulnerability to such socio?economic and health problems among people 133 with mental health disorders I. 'i i 1 134 These results suggest that broader prevention outside of clinical settings is necessary. This can Include l35 educating the public to know the warning signs of suicide and how to respond. especially if someone discloses [36 Suicide intent- strengthening economic Supports ie.g., through housing stabilization policies, household ?nancial l3?r support}; creatingprotective environments leg, reducing access to lethal means amongpeople at risk, I33 implementing workplace policies conducive to help?seeking and encouraging overall well?being, and zoning to I39 reduce alcohol outlet density}; teaching coping and problem-soiving skills early in life to foster lifelong healthy I40 relationships and promoting social connectedness to increase one's sense of belonging and access to resources. Ml To address findings that t?two-thirds of people with MHP had a history of any treatment ntental?healthaandyler I42 sebstanee?abose?tseasmtant and were in meet?treatment when they died states and 143 communities can. - I44 treatment modalities-1e e..g Isash?as Wgreater I45 boratiye ca there pyl, provide . 146 access to behavioral health providers lespeciallg in underserved as?agd encourage greateris I47 iphysical and behavioral health that?beeter?se-ppeet?semde 143 healthcare systems aring for l6? friends/family offer a suicide has taken place and 163 safe media reporting and messaging in the aftermath of a suicide prevent 164 suicide contagion {12) and keep people safe. Such a comprehensive approach to suicide prevention to support 165 all people with or without mental health problems is essential andS_some states. such as Colorado, are 166 underway planning te?implement?such a?eernpsehensieea strategy?appeeaeh se?soieide?peeveneen?HB}. Comment True pointbut it .I- of, a .- a ICom moot [2am]: updated suggests at lea st to me there is limited prevention opportunity. Given 22% disclosed suicide intent, perhaps there needs to be more stated here. Idea: suicides can occur with minimal deliberation but many individuals disclose suicidal intent or demonstrate other risk factors that help in identifying at-risk individual who need immediate acoess to Lsupportive services. 2 Comment Wonder if it might be good to move this point down into the paragraph about comprehensive strategies or its own paragraph since it is applicable to those with known and not known MPH. If moved out of this paragraph specific to MPH, and framed more broadly about what all persons need, then attention to a couple of vulnerable populations per earlier Lcomrnent could be added to this idea. 1 Comment [2:119]: Prefer to keep this as is because we want to want to radvocate' for better treatment and services for Lpeople with MHP. Comment I like the evolution I have seen of the results and discussion to emphasize the difficulties of those without known MHP. I do think it?s important as done for the discussion to talk speci?cally about those with known MPH. But, the two MPH group paragraphs in the discussion seem to tip the balance more toward the MPH group. I suggest JL weaving into the previous paragraph that 167 168 169 170 171 172 173 174 175 I76 177 178 179 180 181 182 183 184 185 186 I87 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.28.18, e-clearance These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next?of-kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist lt is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem with associated risk factors?aee?available. States and communities can data to better understand their suicide problem and can use resources such as Preventing Suicide: a Technical Package of Poiicies, Programs, and Practices (12) to evidence- based comprehensive suicide prevention to save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARSI. Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2013. In. 2. Ivey?Stephenson A2, Crosby AE, iack SPD, Haileyesus T, Kresnow-Sedacca M1. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001- 2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy 5, Xu 1, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. foice of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and ofthe National Action Alliance for Suicide Prevention. 2012. 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 Rev 3.28.13, e-clearance 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 2010;1411124-34. 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis 5P, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017zinjuryprev-2017-042366. 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 2013 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_5uicide Vital Signs MMWR 3.27.18 (table 1.fig 1} Stone_5uicide Vital Signs MMWR 3.22.13 (Tables 2&3} Word Count: Rev 3.23.18, e-clearance Comment IFCH: I think some sort of i clarification is needed to distinguish from previous two sentences since what is 3 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, _l reported here are my data on fu? examined time period or all 50 states. 4 lVlS:1 Kristin lv'l. Holland, Asha Z. Ivey-Stephenson, PhD;1Alex E. Crosby, M01 Tracked an idea but there are likely other 5 Background: Suicide rates in the United States have risen nearly 30% since 1999. Meetal?health?pseblems?(MHP?l ways to Clarify 6 state?level trends ln,?suieide and ether?the multiple {Comment Updated 7? contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Short Title: Vital Signs: Increasing Trends in State Suicide 2 Rates and Exploring Suicide's Multiple Contributing Circumstances num- 8 Methods: Trends in age?adjusted suicide rates among people aged :10 years, by state and sex, across six 5 9 consecutive three?year periods (1999?2016}, were assessed using data from the National Vital Statistics System i Ill for 50 states and Washington, D.C. Data from the National Violent Death Reporting System covering ,i 2? states in 2015, were used to examine contributing circumstances among decedents with and without known l2 M?HPmental health probiems l3 Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases I4 of more than 30%. Rates increased significantly among males and females In 34 and 43 states, respectively. ii I IS .ri not have a known Among decedents If) with circumstance information, several circumstances,?fnciudfri? relationship problems/loss {45.1% vs life stressors (54.2% vs 49.2%] and recentfimpending crises (32.9% vs were significantly more likely 13 among those without a known than decedents with MHP, but were common across groups. l9 Conclusions: Suicide rates increased significantly across most states from 1999-2015. Various circumstances 20 contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach 2 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 22 23 after a suicide occurs. 24 INTRODUCTION 25 BACKGROUND AND PURPOSE 26 In 2015, nearly 45,000 suicides [age-adjusted? occurred in the United States among people I 2? 210 years old Between 1999 and 2015, suicide rates increased across ?sexes, raciallethnic groups, and 28 urbanization levels (2, Suicide is the 10th leading cause of death and is one of just three leading causes that 29 are increasing Additionally. rates of Emergency Department visits for nonfatal self-harm, a key risk factor 30 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 3 nation more than $69 billion in direct medical and work loss costs Ill. 32 The Notional Strotegyfor Suicide PreventioanSSP} calls for a public health approach to suicide prevention 33 with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 34 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 35 Despite the NSSP guidance, suicide prevention efforts largely focus on 36 interventionsmentai?healeh focused on people with mental health problems or pgoplg who have attempted suicide (I5). Other 3? - - .-- 38 contributing circumstances include social isolation, relat 39 lethal means substances, firearms, bridges) among people at risk, and poor coping and problem?solving 40 Expanded awareness of thes_e additional circumstances that?contributl?ge 1 ionship problemsgancl economic problems, access to Fiev 3.23.18, e-clearance 4 to suicide risk apart?feam?Mluliirand action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-speci?c trends in suicide 42 43 rates, assesses the multiple contributing factors to guicide, and provides recommendations for multi-level 44 comprehensive suicide prevention. 45 METHODS 46 Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in younger children 4? are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death 48 certificate records {international ClosSificotion of Diseases 10?? Revision, underlying-cause-of death codes K60- 49 X84, 003}. Age-specific population estimates were obtained from US. Census Bureaulli-lational Center for 50 Health Statistics bridged-race population data releases. SI 52 National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods 53 spanning 1999?2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed 54 per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?year data 55 aggregates, employing weighted least squares regression with inverse?variance weighting. Modeled rate trends 56 are reported in terms of average annual percentage changes 5? 53 Characteristics?nd circumstances of suicide decedents 210 years, with and without known I'leP, were '1 Comment Suggest pulling out the references to tables 2 and 3 since table 1 hasn't been mentioned yet and helps saves on word count. lfyou want to keep them in, suggest adding ?table 1? somewhere into the above paragraph. Comment Iaal?JI: Took the ref to tables 1-i- Comment Given attention to this 5 particular at~risk population in the results, made me wonder if some mention of needs of this group should be in the ,i discussion. There are other higher risk 5 groups described below. A consideration i could be in the discussion about I prevention needs of underserved areas {bottom of page 3} some phrasing could he added about the need for services for 5 higher risk groups, such as those who have served in the military and those who have difficulties, such as MPH, intimate partner problems, have physical health 59 compared in the 27 states with complete data participating in National Violent Death Reporting System 60 in 2015. MUD-RS defines MHP as disorders and listed in the Diagnostic and Statistical Manual 6] of Mental Disorders, Fifth Edition with the exception ofalcohol and other substance dependence, which are 62 captured separately in NUDRS. aggregates data from three primary data sources: death certificates, 63 coroner/medical examiner reports {including toxicology}, and law enforcement reports. Decedents with and 64 without known were compared using Chi?square tests. Logistic regression analyses estimated adjusted 65 odds ratios with 95% confidence intervals controlling for age group, sex, and race/ethnicity. 66 RESULTS 67 The most recent overall suicide rates {representing 2014?2015} varied four?fold, from 5.9 to 29.2 68 {Montana} per 100,000 persons per year {Table Across the study period, rates increased in all states, except 69 Nevada {which had a consistently high rate throughout}, with absolute increases ranging from +0.3 {Delaware} to +8.1 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% {Delaware} to {North ?i?l Dakota}, with increases of more than 30% observed in 25 states {Table 1, Figure T2 ?3 Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females {43 ?4 states}, as well as for the US. overall {Table Nationally, the model-estimated AAPC forthe overall suicide T5 rate was By sex, estimated national rate trends further indicated significant increases for males 76 and females problems, and have ?nancial and legal dif?culties. Or on page 4 where "identifying and better supporting people Lat risk? is stated an e.g. could be added. Comment [21:19}: Did not end up adding this as it seemed odd to call out this one Lpopulation. 7?7 7'8 Suicide decedents without known MHP were compared to those with MHP in 2? states. T9 While all decedents were predominately male {Table 2; 76.8%} and non-Hispanic white those without 80 known MHP, relative to those with MHP, were more likely male {83.6% vs. 63.8%; odds ratio 95% CI 2 8 2.2?2.5} and racialfethnic minorities range: Suicide decedents without known MHP also had 82 significantly greater odds of perpetrating homicide?suicide {adjusted odds ratio 2.9, 95% CI 83 Among adult decedents, 20.1% and 15.3% of people without and with respectively, ever served, or were 34 currently Serving, in the 0.5. milita ll} [04 l05 lilo ll]? IOS Illll'ir IIS ll?EiII IZU Ill [22 123 l24 125 Rev 3.23.18, e-clearance while firearms were the most common method of suicide overall and for both groups, decedents Comment Well: It appearg without known l'leP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by x" being compared to non-MHP. Tracked {25.9% vs 31.3%] or poisoning [10.4% vs 19.3%) than those with known what i think the Edit i5 bl-li PIEEISE dOUblE These differences remained significant in the adjusted models. Chad? e? . . Comment [21119]: You are correct. Should Decedents Without known MHP were less likely to receive toxicology testing. Among those Witl'l toxicology be with in the second instances not results, decedents without known MHP were less likely to test positive for any substance overall 95% without. but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP [N=9,357l had available circumstances information {Table People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known l'leP versus those with known MHP had significantly greater likelihood of any relationship problemfloss [45.1% vs. intimate partner problems (30.2% vs. [115% vs. and recently perpetrating interpersonal violence vs. They Were also more likely to have experienced any life stressors [54.2% vs such as criminal-legal problems vs. or evictionfloss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. Decedents without known MHP had signi?cantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be released from a correctional facility [25.7% vs. or hospital (43.3% vs. than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group Were released from facilities. Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation [23.0% vs. 40.3%] and prior suicide attempts vs. More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-1015, 44 states saw significant soieide-rateincreasesincreases in suicide rates and 25 states experienced rate increases of more than 30%. Rates increased significantly far?among males, {in 34 states}; and fesfemales; [in 43 states}. This finding is consistent with prior research showing a decreasing gender gap in male-female suicide rates between research into the specific causes of these trends is necessary. Fo?unatelyln the meantime ean?shed?ligh-t?on?provides insight into thesuiclde circumstanCES and help states identify prevention regularly state that 1suicide :is not caused by a single factor;? however. suicide prevention research and practicepceventien heavily weighted twir? activities In contrasLtIhe current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship prehIeHas-issues and ethilife stressors such as criminal-legalmastees, housingeeic?tienflessef 3 Rev 3.28.18, e-clearance 26 home, and health problems They also experienced moreand recent?-or: impending crises which may point to 127 suicides took place man+soieides?aod 123 minimal deliberation In addition to educatingabout how to l29 disclosures of suicide risk, broader prevention to address non?clinical risks is critical and can include 130 strengthening economic supports Ie. g. through housing stabilization policies household financial support); l3] sigirlig?egrly_ healthy Ielationships oyer the life course; laud I32 promoting social connectedness to increase one 5 sense of belonging I 133 The above strategies are aISo applicable to Amongpeople with MHP who, in the current study struggled more 134 witlljobf?nancial problem; bereavement issues, and family crises, and who commonly experienced I35 argumentsi'conflicts, intimate partner, and health problems. These ?ndings support a body of literature I36 indicating increased vulnerability among people with MHP to socIo-economic and health problems I11). In the 137T c_ inical pfpeople with IijP had a history of _a ny ISIS abuse-treatment and everhalt??ti?is were in current-treatment when they died. This suggests the need for that 139 - -- b_roader I40 implementation of indence- based treatment medaiities?Ie e.g soeh?asooetor-patieot collaborative careLvmodels-end cognitive- -behavioral therapyL: access to behavioral health 3" I42 providers Iespecially' In underserved areaslisrneetled, and expanded I43 integraticme oiphysical and behavioral health I44 healthcare systems, especially through care transitions l' I l4? 148 [49 In addition to the abovementioned stragegjgs?her ISO underseorethe impoetanceot- thateddress I55 strategies?includd; creating protective environments [56 bS?i'EgsrrEdUCing access to lethal means Ie. g. medications firearms} among people at risk, creating 15? policies toencouragingpromotee employee help? seeking and connectedness,? 158 - - -- ,and zoningto reduce [59 alcohol- outlet density; better identifying people at risk; supporting after a suicide has taken place 160 and safe reporting by the media' In order to prevent suicide contagion Some states, such as Colorado, are planning to?Hm-plement?such a comprehenswe [62 approach to suicide prevention [63 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, 164 might have been impacted by large proportions of injury deaths of undetermined Intent [potentially biasing 165 reported suicide rates downward], or decreased percentages of such deaths over time {potentially biasing I-s-s estimated rate trends upward]. Second, is not yet nationally representative; the 2? states included l6?? represent 49.6% of the population 168? 4 Comment True point but it suggests at lea st to me there is limited prevention opportunity. Given 22% disclosed suicide intent, perhaps there needs to be more stated here. Idea: suicides can occur with minimal deliberation but many individuals disclose suicidal intent or demonstrate other risk factors that help in identifying at-rislc individual who need immediate acoess to Lsupportive services. [Corn merit [cam]: updated Comment II'II'onder if it might be good to move this point down into the paragraph about comprehensive strategies or its own paragraph since it is applicable to those with known and not known MPH. If mOVed out of this paragraph specific to MPH, and framed more broadly about what all persons need, then attention to a couple of vulnerable populations per earlier Lcomment could be added to this idea. 1r I I because we want to want to 'advocate' for better treatment and services for Lpeople with MHP. Comment I like the evolution I emphasize the difficulties of those without known MHP. I do think it?s important as done for the discussion to talk speci?cally about those with known in the discussion seem to tip the balance more toward the MPH group. I suggest 3 Comment [2:119]: Prefer to keep this as is have seen of the results and discussion to MPH. But, the two MPH group paragraphs .1 1 weaving into the previous paragraph that a a . 169 120 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191. 192 193 1 94 195 196 197r 198 199 200 201 202 203 204 205 206 207 208 209 210 2] 1 212 213 Rev 3.28.13, e-cleara nce Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent ofinformant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next?of?kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem with associated suieide risk States and communities can data to better understand their suicide problem and can use resources such as (2005 Preventing Suicide: a Technical Package of Policies, Programs, and Practices (12) to evidence based comprehensive suicide prevention to save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstane3@cdc.gov 220?438?3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, IiDivision of Analysis, Research, and Practice Integration, National ILit-enter for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web?based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 20011 2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics 201?. 5. Office ofthe Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 3. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders (DEM-SP): American Pub; 2013. 214 215 216 217 218 219 221 222 223 224 225 226 227r 228 229 Rev 3.28.13, e-clearance 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, D'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholovv EN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017zinjuryprev-2017-042366. l4. Milner A, SveticicJ, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.22.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 3.22.18 (Tables 2&3} Word Count: Rev 3.28.18, e-clearance Comment IFCH: I think some sort of clarification is needed to distinguish from previous two sentences since what is Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, PhDr? Keming Yuan, a i 3- Kristin M. Holland, PhD;1Asha Z. Nev-Stephenson, Ales E. Crosby, 1'le1 reported here are "at data examined time period or all 50 states. 4 Background. rates In the United States have risen nearly 30% since 1999. Mental health problems i Tracked an idea but there are likely other 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing ways to Clarify circumstances can inform comprehensive state suicide prevention planning. 7 Methods.- Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across sin ti consecutive three-year periods {1999?2016}, were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 27 states Ill in 2015, were used to examine contributing circumstances among decedents with and without known MHP. II Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states euperiencing increases [2 of more than 30%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. l3 @nong decedents in 2015 in 2? states, nger half of decedents did not have a known MHP. Among . 14 dependents-with circumstance information, IS life stressors [54.2% vs 49.2%) and recentfimpending crises [32.9% vs were significantly more If) likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999?2016. Various circumstances IS contributed to suicides among people with and without known 19 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 20 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 21 after a suicide occurs. 22 INTRODUCTION 23 BACKGROUND AND PURPOSE 24 in 2016, nearly 45,000 suicides [15.6f100,000 [age?adjusted? occurred in the United States among people 25 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 26 urbanization levels Suicide is the 10?? leading cause of death and is one ofjust three leading causes that 2? are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm. a key risk factor 28 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 29 nation more than $59 billion in direct medical and work loss costs 30 The National StrategyforSuicide PreventioanSSP} calls for a public health approach to suicide prevention 3 with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 32 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 33 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 34 mental health problems or providing follow-up care to people who have attempted suicide Other 35 contributing circumstances inciude social and economic problems, access to lethal means substances, 3o firearms, bridges} among people at risk, poor coping and problem?solving skills, and prior suicide attempts (Si. Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action 38 to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in 39 achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing 40 factors, and provides recommendations for multi-Ievel comprehensive suicide prevention. 1 Rev 3.23.18, e-clearance 4 METHOD5 1 Comment Suggest pulling out the . . . . . . . . references to tables 2 and 3 since table 1 42 SUICICIE rates were analyzed for people aged :10 years only, as attributions of suucidal Intent in younger children ,i hasn't been mentioned yet and helps 43 are variable Age?speci?c suicide counts were tabulated based on National Vital Statistics System coded death i saves on word count. ?you want to keep 4-4 certi?cate records [international Classrficotion of Diseases 10'? Revision, underlying-cause?of death codes K60- them in; suggest adding "tabie 1? ,somewhere into the above paragraph. i 45 X84, YBTO, Age-specific population estimates were obtained from U.S. Census Bureauf'll-lational Center for ,i 1 4? ,i Comment Given attention to this . . . . . i particular at?risk population in the results 48 National and state-level rate estimates were calculated for so: consecutive three-year aggregate periods 5 made me wonder if some mention of I 49 spanning 1999-2016. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed 5 needs at this group should be in the 50 per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three~year data ,i discussion. There are other higher risk groups described below. A consideration aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 46 Health Statistics bridged~race population data releases. could be in the discussion about 5 . . 52 are reported in terms of average annual percentage changes 53 i prevention needs of underserved areas I 54- Characteristics ?ialale?Etand circumstances-Fables} of suicide decedents 210 years, with and without knoWn 4? i {bottom ?f page 3i 59m? phrasing ?Did be added about the need for services for higher risk groups, such as those who 55 were compared in the 2? states with complete data participating in National Violent Death 56 Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and have served in the militaryand those who 5? Statistical Manual of Mental Disorders, Fifth Edition with the Exception of alcohol and other substance 58 dependence, which are captured separately in aggregates data from three primary data sources 59 death certificates, coroner/medical examiner reports [including toxicology}, and law enforcement difficulties. Or on page 4 where ?identifying and better sopporting people i have difficulties, such as MPH, intimate i 60 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression i i get risk" is stated an e.g. could be addedpartner problems, have physical health problems, and have financial and legal analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and Comment It appears non-MHPis being compared to non-MHP. Tracked whatl think the edit is but please double check. is 152 race;l ethnicity. 63 RESULTS at The most recent overall suicide rates [representing 2014?2015} varied four?fold, from 5.9 to 29.2 65 {Montana} per 100,000 persons per year (Table Across the study period, rates increased in all states, except as Nevada [which had a consistently high rate throughout}, with absolute increases ranging from +0.3 {Delaware} or to +8.1 {Wyoming} per 100,000. Percentage increases in rates ranged Ifrom +53% {Delaware} to [North 68 Dakota}, with increases of more than 30% observed in 25 states [Table 1, Figure 69 i0 Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females [43 TI states}, as well as for the US, overall {Table Nationally, the model~estirnated AAPC for the overall suicide 7?2 rate was By sex, estimated national rate trends further indicated significant increases for males T3 and females 7?4 7'5 Suicide decedents without known MHP lN=11,039} were compared to those with lv'lHP in 2? states. as While all decedents were predominately male {Table 2; i6.8%} and non-Hispanic white those without Ti known MHP, relative to those with l'leP, were more likely male {33.6% vs. 68.8%; odds ratio 95% CI 78 2.2?2.5} and raciaifethnic minorities (0R range: Suicide decadents without known MHP also had i9 significantly greater odds of perpetrating homicide?suicide {adjusted odds ratio 2.9, 95% CI 80 among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were til currently serving, in the 0.5. militarv? 82 While firearms were the most common method of suicide overall and for both groups, decedents 83 without known MHP were more likely to die by firearm {55.3% vs. 40.6%} and less likely to die by I 84- {25.9% vs 31.3%} or poisoning {10.4% vs 19.3%} than those without known i i i 85 differences remained signi?cant in the adjusted models. 2 Ftev 3.23.18, e-clearance 86 Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology 8? results. decedents without known l'leF were less likely to test positive for any substance overall 95% 83 but more likely to test positive for alcohol laClFt=1.2, 95% 89 All suicide decedents with MHP lN=9,40Tl and approximately 85% without known lv'lHF' had available 90 circumstances information {Table 3). People without known MHP were less likely to have any substance abuse 9 problems 95% While two-thirds of those with known MHP had a history of mental health 02 or substance abuse treatment jUSt over half were in current treatment. 93 Decedents without known MHP versus those with known MHP had significantly greater likelihood of any 04 relationship problemfloss (45.1% vs. specifically intimate partner problems [30.2% vs. 05 [115% vs. and recently perpetrating interpersonal violence vs. They or, were also more likely to have experienced any life stressors [54.2% vs 491%], such as criminal?legal problems 0? {101% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 98 the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the 99 adjusted models. Among those with crises, intimate partner and physical health problems were the most l0(i common types for both groups and did not differ between them. IUI Decedents without known MHP had signi?cantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be l03 released from a correctional facility [253% vs. or hospital l43.?% vs. 33.0%] than those with a known I04 MHP. Among decedents with known lv'lHP who were recently released from an institution 463% of this l05 group were released from facilities. lilo Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of I07 suicidal ideation [23.0% vs. 40.8%] and prior suicide attempts {10.3% vs. More than 1 in five people in [08 both groups disclosed suicide intent [22.4% vs. lil9 Conclusions and Comments It} From 1999?2015, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 ?2 gap in male-female suicide rates between 1999?2014 Overall, half of the states experienced substantial ?3 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is [5 contributed to recent suicides, and can help guide prevention activities. If) Researchers and practitioners regularly state that ?suicide is not caused by a single factor;' however, research and prevention practices often focus on identifying and treating lv'lHP The current study found that more 118 than half of suicide decedents in did not have a known l'v'lHP. This group suffered more relationship I I9 problems and life stressors such as criminal?legal matters, evictionfloss of home, and recent or impending crises. 20 This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal l2l deliberation (10 [23 half were in current treatment. This suggests that additional supports for this population are needed to keep 124 them safe. This includes broader implementation of affordable and effective treatment modalities such as 126 behavioral health providers in underserved areas is needed, as is expansion of healthcare systems to integrate 3 Comment True point but it suggests at least to me there is limited prevention opportunity. Given 22% disclosed suicide intent, perhaps there needs to be more stated here. Idea: suicides can occur with minimal factors that help in identifying at~risk LsuFIportive services. deliberation but many individuals disclose suicidal intent or demonstrate other risk individual who need immediate access to states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender ?4 necessary. Fortunately, data from the 27 states participating in can shed light on the circumstances that Comment Wonder if it might be good to move this point down into the paragraph about comprehensive strategies or Its own paragraph since it is applicable to those with known and not known MPH. If moved out of this paragraph specific to MPH, and framed more broadly about what all persons need, then attention to a couple of vulnerable populations per earlier Lcomment could be added to this idea. I28 I29 3fl I31 I32 [34 I35 I36 13? I33 I40 I4I I42 I43 I44 I45 I46 I43 I4?:ll I50 I52 153 I54 155 I56 158 I59 I?l I62 I63 I64 I135 166 lb? I158 I69 Rev 3.23.18, e-clearance physical and behavioral health that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as joblfinancial, health problems. These associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor social, health, and economic outcomes Together, these results underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports leg, housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support: and identifying and better supporting people at risk. Other strategies include creating protective environments leg, reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined Intent [potentially biasing reported suicide rates downward], or decreased percentages of such deaths over time [potentially biasing estimated rate trends upward]. Second, is not yet nationally representative; the 2? states included represent 49.6% thhe population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next?of?kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent lv'lHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can Use data from and resources such as CDC's Preventing Suicide: Technical Package of Policies, Programs, and Practices {11} to better understand theirsuicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Con?ict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, Author Affiliations: Comment IFCII: I like the evolution I have seen of the results and discussion to emphasize the dif?culties of those without known MHP. I do think it?s important as done for the discussion to talk speci?cally about those with known MPH. But, the two MPH group paragraphs in the dlscussion seem to tip the balance more toward the MPH group. I suggest weaving into the previous paragraph that those with known had other dif?culties rather than this being a standalone paragraph. Aiso consider If the prevention implication described here is also applicable to those without MPH. There seemed to be a gap in discussion paragraph #2 where similar precipitating circumstances are noted for the non-M HP group but no prevention strategies are presented. Prevantion strategies for these types of risk factor could be described either in the non?MPH paragraph or this paragraph cold be framed more broadly about the prevention needs of both non? LMPH and i'leH groups. 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.28.18, e-clearance 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. A. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et aI. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives ofSuicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, rroII PW. Characteristics of impulsive suicide attempts and attempters. Suicide and LifeAThreatening Behavior 11. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 12. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev~2017~042366. 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1] Stone_Suicide Vital Signs MMWR 3.27.18 (Tables 28:3} Word Count: 185871800 OJ woo-4Rev 4.5.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedii occurred in the United States among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10"h leading cause of death and is one ofjust three leading causes that are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The National Strotegyfor Suicide Prevention{NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts are largely clinically-oriented, focused on treating mental health problems and preventing re?attempts (6). Apart from MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor coping and problemrsolving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes Rev 4.5.18, e-clearance state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi?Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Classification of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS [665] The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicideesuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by Fiev 4.5.13, e-clearance R4 [26.9% vs 31.3%) or poisoning [10.4% vs 19.8%) than those with known I. Comment i would add in the ?li: 85 These differences remained significant in the adjusted models. it for opioids as it is timely and relevant 36 Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology if commit!? lthGUEht they were 87? results, decedents without known l'vll-lP were less likely to test positive for any substance overall 95% like?? to be from other I facility which Included drug rehab? E48 but more likely to test positive for alcohol [aGFi=1.2. 95% [59 All suicide decedents with lleP and approximately 85% without ,I'r 9i} circumstances information [Table People without known MHP were less likely to have any substance abuse 5 9 problems 95% While two-thirds of those with known MHP had a history of mental health Ii 92 or substance abuse treatment just over half were in current treatment. 5 93 Decedents without, versus those with, known MHP, had significantly greater likelihood of any relationship 94 problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. ,5 95 [12.5% and recently perpetratlng interpersonal violence vs. They also were more likely 96 to have experienced any life stressors [54.2% vs such as criminal~legal problems [10.2% vs. or ll 9? evictionfloss of home vs. and were more likely to have had a crisis within the preceding or 98 upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. 99 Among all people with crises, intimate partner problems were the most common types and did not differ by IUD group. Similarly, among people without versus with MHP, physical health problems {23.2% and 21.4%] and I i i IOI job/financial problems [15.5% and 16.3%) were commonly experienced and did not differ by group. fi2 Decedents without known MHP had signi?cantly lower odds of recent release from any institution MB 95% out among those who were recently released they were significantly more likely to be Iii-4 released from a correctional facility [25.2% vs. or hospital vs. 33.0%] than those with a known [on group were released from facilities. Ill".F Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of [lift suicidal ideation [23.0% vs. 40.3%] and prior suicide attempts vs. More than one in five people in both groups disclosed suicide intent [22.4% vs. I Ill Conclusions and Comments From 1999-2015, 44 states saw significant increases in suicide rates and 25 states experienced substantial l2 increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and l3 females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- I [4 female suicide rates between 1999-2014 Additional research into the specific causes ofthese trends is [5 necessary. Fortunately, data from the 22 states participating in provides important insight into suicide no circumstances and can help states identify prevention priorities. ll? Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention research and practice is heavily oriented towards treatment of lv'lHP and prevention of reattempts. ?9 Additional focus on non?mental health factors, further upstream, is essential to a public health approach as l2ll the current study found that more than half of suicide decedents in did not have a knowh MHP. This l2l group suffered more from relationship problems and other life stressors such as criminal?legal matters, 122 evictionfloss of home, and recent or impending crises. l23 Similarly, people with lleP also experienced other life stressors such relationship, andfor 24 physical health problems. These findings point to the need to both prevent the Conditions associated with 3 125 126 127 128 129 131} 131 132 133 134 135 .136 137 138 139 140 141 142 143 145 146 147V 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 Rev 4.5.13, e-clearance mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two-thirds of this group had a history of any mental health and/or substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor?patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physical/mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward), or decreased percentages of such deaths overtime (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included rapresent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Paiicies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. 4 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 1% 197 198 199 200 201 202 203 204 205 206 207 Rev 4.5.13, e-clearance Corresponding author: Deborah M. Stone, 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atia nta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca Ml. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy 5, Xu 1, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wesserman D, van Heeringen K, Arensman E, Sarchia pone M, et 31. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 201?. 3. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed 1, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 2005;361511491-510. Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.5.18 (Tables 2&3} Word Count: 186871800 Rev 316,18, e-clearance Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates Comment lH-Hli Is there a more up?to- date US?based reference than Rosenman 1993? Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. kegler, Keming Yuan, 2 3- Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Phi};1 Alex E. Crosby, l'v'lD1 4 Background: Suicide rates in the United StatEs have risen nearly 30% since 1999. Mental health problems 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing a circumstances can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six ii consecutive three-year periods {1999?2016}, were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 27 states Ill]I in 2015, were used to examine contributing circumstances among decedents with and without known II Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases [2 of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Over half of decedents did not have a known Among decedents with circumstance information, 14 several circumstances, including relationship problems/loss {45.1% vs life stressors {54.2% vs 491%} and IS recent/impending crises (32.9% vs were significantly more likely among those without a known l'leP [6 than decedents with MHP, but were common across groups. [7 Conclusions: Suicide rates increased significantly across most states from 1999?2015. Various circumstances I?d contributed to suicides among people with and without known 19 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 20 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 21 after a suicide occurs. 22 INTRODUCTION 23 BACKGROUND AND PURPOSE 24 in 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States among people 25 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 2f) urbanization levels Suicide is the 10?? leading cause of death and is one ofjust three leading causes that 2? are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm. a key risk factor 28 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 29 nation more than $59 billion in direct medical and work loss costs The Notional StmtegvforSuicide PreventioanSSP} calls for a public health approach to suicide prevention i 3 with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 32 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. i 33 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 35 access to lethal means leg, substances, firearms, bridges] among people at risk, poor coping and problem- 36 solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from and action to address them, can help reach the nation?s goal of 3'8 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific 39 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level 40 comprehensive suicide prevention. Rev 3.26.18, e-clearance METHODS Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in younger children are variable (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Classification of Diseases 10?r Revision, underlying-cause-of death codes X60- X84, Y87.0, U03). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridged-race population data releases. National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCS). Characteristics (Table 2) and circumstances (Table 3) of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception ofalcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and race/ethnicity. RESULTS The most recent overall suicide rates (representing 2014-2016) varied four-fold, from 6.9 (D.C.) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +59% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states) and females (43 states], as well as for the U.S. overall (Table 1). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%) and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5) and racial/ethnic minorities (OR range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%} and less likely to die by (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those without known MHP. These differences remained significant in the adjusted models. Fiev 3.26.18, e?clearance E46 Decedents without known MHP Were less likely to receive toxicology testing. Among those with toxicology E47 results, decedents without known lleF were less likely to test positive for any substance overall 95% SS but more likely to test positive for alcohol laClR=1.2, 95% 89 All suicide decedents with MHP and approximately 85% without known lv'lHF' had available 90 circumstances information {Table 3). People without known MHP were less likely to have any substance abuse 9 problems laDFt=U.7, 95% While two-thirds of those with known MHP had a history of mental health 92 or substance abuse treatment just over half were in current treatment. 93 Decedents without known MHP versus those with known MHP had significantly greater likelihood of any 94 relationship problem/loss [45.1% vs. specifically intimate partner problems [30.2% vs. 95 [115% vs. and recently perpetrating interpersonal violence vs. They 96 were also more likely to have experienced any life stressors [54.2% vs such as criminal?legal problems 9? {101% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 98 the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the 99 adjusted models. Among those with crises, intimate partner and physical health problems were the most IOU common types for both groups and did not differ between them. Decedents without known MHP had signi?cantly lower odds of recent release from any institution IUZ 95% but among those who were recently released they were significantly more likely to be released from a correctional facility [253% vs. or hospital vs. 33.0%) than those with a known Ill-4 MHP. Among decedents with known lv'iHP who were recently released from an institution 45.7% of this group were released from facilities. lilo Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of If)? suicidal ideation [23.0% vs. 40.8%) and prior suicide attempt {10.3% vs. More than 1 in five people in [(18 both groups disclosed suicide intent [22.4% vs. lilii? Conclusions and Comments Ilfl From 1999?21115, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender ?2 gap in male-female suicide rates between half of the states experienced substantial ?3 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is ?4 necessary. Fortunately, data from the 27 states participating in can shed light on the circumstances that 15 contributed to recent suicides, and can help guide prevention activities. I I6 Researchers and practitioners regularly state that ?suicide is not caused by a single factor;' however, research I l? and prevention practices, almost solely, focus on identifying and treating current study found that 118 more than half of suicide decedents in did not have a known l'leF'. This group suffered more relationship I I9 problems and life stressors such as criminalrlegal matters, evictionfloss of home, and recent or impending crises. IZO This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal l2] deliberation (10). l22 Among people with MHP, twovthirds had a history of mental health andfor substance abuse treatment and over [23 half were in current treatment. This suggests that additional supports for this population are needed to keep 124 them safe. This includes broader implementation of affordable and effective treatment modalities such as 125 doctor-patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to 126 behavioral health providers in underserved areas is needed, as is Expansion of healthcare systems to Integrate 3 .- ?fty?- Add citations to .. support this statement. Comment zaf9): Alex? add Zalsman again? Don'thave room for another ref. Rev 3.26.13, e-clearance physical and behavioral health that better support suicide prevention and patient safety, especially through care transitions (11). Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, relationship, andi'or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor social, health, and economic outcomes Together, these results underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk. Other strategies include Creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time [potentially biasing estimated rate trends Upward}. Second, is not yet nationally representative; the 27 states included represent 49.5% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: Technicoi Package of Poiicies, Programs, and Practices (11) to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gg 770438?3942 Author Affiliations: 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.26.18, e-cleara nce 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISOARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 2017;66l18jzl-1E. 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. A. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Rosenman SJ. Preventing suicide: what will work and what will not. The Medical Journal of Australia 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life?Threatening Behavior 11. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 12. World Health Organization. Risks to mental health: An overview ofvulnerabilities and risk factors. Geneva: WHO 2012. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 20172injuryprev+2012-042366. 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 510. Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1] Stone_Suicide Vital Signs MMWR 3.27.18 (Tables 28:3} Word Count: 1861/? 1800 Rev 3.26.18, e-clearance 2 3 circumstances can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six ii consecutive three-year periods (1999-2016}, were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, DC. Data from the National Violent Death Reporting System, covering 27 states l0 in 2015, were used to examine contributing circumstances among decedents with and without known [6 than decedents with MHP, but were common across groups. [7 Conclusions: Suicide rates increased significantly across most states from 1999?2015. Various circumstances IS contributed to suicides among people with and without known l9 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 20 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 2 after a suicide occurs. 22 INTRODUCTION 23 AND PURPOSE 24 in 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States among people 25 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 2f) urbanization levels Suicide is the 10'h leading cause of death and is one ofjust three leading causes that 2? are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm. a key risk factor 28 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 2?9 nation more than 569 billion in direct medical and Work loss costs 30 The Notional Strategyforsmcide PreventioanSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 32 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 33 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 34 mental health problems contributing circumstances include social and economic problems, 35 access to lethal means among people at risk, poor coping and problem- 36 solving skills, and prior suicide attempts i5}. Expanded awareness of the additional circumstances that .32 contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of 3'8 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific 39 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level 3 40 comprehensive suicide prevention. Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, M59 Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 4 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states esperiencing increases [2 of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, 14 several circumstances, including relationship problems/loss {45.1% vs life stressors [54.2% vs 49.2%} and IS recentfimpending crises (32.9% vs were significantly more likely among those without a known MHP Comment IHJH: Is there a more up?to- date US?based reference than Rosenman 1993? Comment [2:119]: Alex, what about the Zalsman systematic review? if you look at ll all the treatment strategies they appear to outweigh any other strategies and if within the limiting access to lethal means, Ll many of the studies are international. ii 1. Zalsman. G., et al., Suicide ll prevention strategies revisited: matrear systematic review. The Lancet 5! 2015.317}: p. sac-ass. i. ii ==adle::gg= La not.vd-a-Ih-ICIG-Iai-tlnl =Il Cal Rev 3.25.18, e-clearance 4 METHODS 42 Suicide rates were analyzed for people aged :10 years only, as attributions of suicidal intent in younger children 43 are variable Age?speci?c suicide counts were tabulated based on National Vital Statistics System coded death 44 certi?cate records [international Classification of Diseases 10'? Revision, underlying-cause?of death codes K60- 45 X84, YETO, Age-specific population estimates were obtained from U.S. Census Bureauf'li-lational Center for 4s Health Statistics bridged~race population data releases. 4? 48 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 49 spanning 1999-2016. Hate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed 50 per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three~year data 5] aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 52 are reported in terms of average annual percentage changes 53 54 Characteristics [Table 2} and circumstances {Table 3} of suicide decedents 210 years, with and without known 55 l'leP, were compared in the 2? states with complete data participating in National Violent Death 56 Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and 5? Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance 58 dependence, which are captured separately in aggregates data from three primary data sources: 59 death certificates, coroner/medical examiner reports {including toxicology}, and law enforcement 60 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression bl analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and 62 race/ ethnicity. 63 RESULTS The most recent overall suicide rates [representing 2014?2016} varied four?fold, from 5.9 to 29.2 05 {Montana} per 100,000 persons per year (Table Across the study period, rates increased in all states, except I as Nevada consistently has one of the highest suicide Eateiiwith . of absolute increases ranging from +0.8 {Delaware} to +8.1 {Wyoming} per 100,000. Percentage increases in rates 68 ranged from +5.91% {Delaware} to (North Dakota}, with increases of more than 30% observed in 25 states 69 {Table 1, Figure "i0 7 Modeled suicide rate trends indicated significant increases for 44 states, for males [34 states} and females [43 ?2 states}, as well as for the 0.5. overall {Table Nationally, the model-estimated MPC for the overall suicide 73 rate was By sex, estimated national rate trends further indicated significant increases for males 74 and females T5 Suicide decedents without known MHP lN=11,039} were compared to those with MHP iN=9,407} in 2? states. While all decedents were predominately male {Table 2; and non-Hispanic white those without i8 known MHP, relative to those with lleP, were more likely male {33.5% vs. 53.8%; odds ratio 95% CI is 2.2?2.5} and raciali'ethnic minorities (on {pant} - Suicide 80 decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide [adjusted RI odds ratio 2.9, 95% Cl Among adult decedents, 20.1% and 15.3% of people without and with 32 respectively, ever served, or were currently serving, in the 0.5. military. 83 While ?rearms were the most common method of suicide overall and for both groups. decedents 84 without known l'v'lHP were more likely to die by firearm {55.3% vs. 40.6%} and less likely to die by 85 {26.9% vs 31.3%} or poisoning [10.4% vs 19.8%} than those without known 36 MHP- These differences remained significant in the adjusted models. 2 whether this should stay here or be here but realize it may not be conventional. i 13. to 1.0. Please con?rm. correct. in the loiiver bound of the 95% for non?white racial/ethnic groups is 1.0 Lil-lispanic} to 1.6 lAlfAN}. a little differently. For the lower limits 31-355!!- - .. numbers across all LL El's and then did the same for upper limits. l'm guessing you guys have done this before and it's I Laccepted? the Cl range here. The reader can find Lthem in the table. in the upper bound ofthe 95% for non?white racial/ethnic groups is 1.3 Comment Imam: KF, K?r. Actually this above comment about Jeff's Linterpretation. Comment lsee howieffis ?Your call. ., Comment lately]: SK, TS, Opinions as to moved to limitations? I?d prefer to keep it J. [Comment I agree with keeping it here. ., Comment Per Table 2. the range Bits for non-white racial/ethnic groups 5 Comment [n19]: it"i?, KF, Jeff looks to be Comment Per Table 2, the range Comment Iza?]: KF, KY, TS--This looks correct to me. Jeff was looking at the data here he took the smallest and largest Comment I see how this could be confusing. l'rn thinking that we could drop Comment Per Table 2, the range LlHispanic} to 3.1 {Other}. Please con?rm. looks like a typo. Should be thinking about this - have made changes in line with this but talked Keming and we are also olt cutting it if that's better. 1 "l h. Fiev 326.18, e-clearance sextlomment Any comment on 87? Decedents without knoWn MHP were less likely to receive toxicology {resting Among those with toxicology substance causes of death? 88 results, decedents without known lleP were less likely to test positive for any substance overall 95% ?g {Comment [2am]: room Comment lagree about the lack .IL. 89 but more likely to test positive for alcohol laClFi=1.2, 95% 9i] All suicide decedents with MHP and approximately 85% without known lv'lHP had available of room. We also don?t address this in prevention, Let?s see if other reviewers circumstances information {Table People without known lv'iHP were less likely to have any substance abuse 9 92 problems laOFi=lJ.7, 95% While two-thirds of those with known MHP had a history of mental health 93 or substance abuse treatment jost over half were in current treatment. raise {Comment Might be better to 5 report as: 2.311% vs. 40.3% ?l 94 Decedents without known MHP versus those with known MHP had significantly greater likelihood of any 95. relationship problem/loss (45.1% vs. specifically intimate partner problems [30.2% vs. i 96 [115% vs. and recently perpetrating interpersonal violence vs. They ,l 9? were also more likely to have experienced any life stressors [54.2% vs 491%}, such as criminal?legal problems _l 93 {10.1% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 99 the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the [00 adjusted models. Among those with crises, intimate partner and physical health problems were the most {Comment Might be better to report as: 29.4% vs. 10.3% {Comment Izal'?l: KF, KY, tried to preserve aDR?sll Thoughts? Comment I see his point. The ?if: differences are more compelling. This would also make the paragraph more consistent since the last sentence used ?l Comment IFKAU: I?m ok with that because these are the ones where we're speaking about the group that has signi?cantly lower odds. My thoughts are the same in general about keeping vs. ekcluding odds ratios, but willing to be flexible think we?ve got to be in this .45common types for both groups and did not differ between them. l0 IOZ Decedents without known MHP had signi?cantly lower odds of recent release from any institution l03 95% but among those who were recently released they were significantly more likely to be l04 released from a correctional facility [253% vs. or hospital l43.?% vs. 33.0%} than those with a known Ills MHP. Among decedents with known lv'lHP who were recently released from an institution 453% of this ?I'a-a In ass w. ?5 - lilo group were released from facilities. It)? Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation (poems, 95% 95% Morqihan 1 in Lprocessll. r{Cotnn'ient Add citations to [(19 five people in both groups disclosed suicide intent [22.4% vs. support this statement. {Comment Infill: Alex? add Zalsman again?I Don'thave room for another ref. 1 ID Conclusions and Comments i From 1999-2015, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 .l l2 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender ,l I l? I Comment Any comment that the i top ranking states are rural and mainly in ?3 gap in male-female suicide rates between 1999?2014 Overall, half of the states experienced substantial ?4 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is i: necessary. Fortunately, data from the states participating in can shed light on the circumstances that il'l the Northwest: lvlT, Alt, WY, UT llo contributed to recent suicides, and can help guide prevention activities. 5; .l,[Comment In?ll Unfortunately we can; I I'i? It is common parlance in the field that :suicide is not caused by a yll' fly: accommodate the added text [8 single factor;j however, suieideresearch and prevention practices, almost?seleiwlargely focus on identifying and ill .lliI 119 treating current study found that more than half of suicide decedents in did not have a known :li I20 MHP. This group suffered more relationship problems and life stressors such as criminaHegal matters, ,lli' eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that if 22 suggest many suicides and attempts occur with minimal deliberation if l23 Among people with l'leP, two-thirds had a history of mental health and/or substance abuse treatment and over ill [24 half were in current treatment. This suggests that additional supports for this population are needed to keep 5: lb them safe. This includes broader implementation of affordable and effective treatment modalities such as l26 doctor-patient collaborative ca re models and cognitive?behavioral therapy, Additionally, greater access to 127 behavioral health providers inEJnderserve-dhreas is needed, as is expansion of healthcare systems to integrate All 3 Rev 3.26.18, e-clearance Comment about Comment [2am]: SK, sure there's a weakness in there. I inserted language above to state that it consistently has one ofthe highest rates. IER physical and behavioral health that better support suicide prevention and patient safety, especially through care 129 transitions I30 Study findings indicate that people with known also experienced other life stressors such as jobi'flnancial, relationship, andy?or physical health problems. These findings point to the need to both prevent the conditions I32 associated with mental health problems in the first place, and to support people with known MHP to decrease [33 their risk of poor social, health, and economic outcomes Comment I don't think we should call out my here. I think text you provide H4 Together, these results underscore the importance of comprehensive statewide suicide prevention activities above 15 sufficient. 5 i 135 that address multiple factors associated with suicide. Prevention strategies may include: strengthening Lib economic supports leg, housing stabilization policies, household financial support}; teaching coping and 5 problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in 5 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, ,4 i i 138 I39I emotional, and social support; and identifying and better supporting people at risk. Other strategies include I40 creating protective environments leg, reducing access to lethal means among people at risk, creating I I4I organizational and workplace policies to promote help-seeking, easing transitions into and out of work for f: I42 people with MHP and other life challenges], supporting family and friends aftera suicide, and assuring safe if . I43 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning I44 to implement such a comprehensive approach to suicide prevention I46 might have been impacted by large proportions of injury deaths of undetermined Intent [potentially biasing I47 reported suicide rates downward], or decreased percentages of such deaths over time {potentially biasing I48 estimated rate trends upward]. [Second is not yet nationally representative; the 2? states included I4";ll represent 49.6% ofthe population [50 ii I45 These findings have at least three limitations. In the state-level analysis, rankings for four states (MDISI Third, abstractors of data are limited to information contained in investigative reports. Therefore, the l52 extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth I53 interviews with next?of?kin often see greater attributions to mental disorders however many [54 methodological variations across studies exist It is likely that some people without known MHP in the 155 current study were experiencing mental health challenges that were unknown, and hence underreported by key 156 informants. However, any lack of awareness of decedent suggests the importance of addressing the broad I5?r range of contributing circumstances. If?? Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are 159 available. States and communities can use data from and resources such as CDC's Preventing Suicide: a lot} Technical Package of Policies, Programs, and Practices {11} to better understand their suicide problem and l6 prioritize evidence-based comprehensive suicide prevention. [62 Acknowledgments [63 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Io4 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. 1&5 Isa Con?ict of Interest No conflicts of interest were reported. Ia? loll Corresponding author: Deborah M. Stone, i'iD-488-3942 I as I 70 Author Af?liations: [7202 203 204 205 206 El}? 208 209 2 I 0 Rev 3.25.18, e-clearance 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis. Research, and Practice Integration, National Center for Injury Prevention and Control, CDC Preference? 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Ctuery and Reporting System i Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Way-Stephenson A2, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001? 2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2015. 4. Kochanek K, Murphy S, In J, Arias E. Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics 201?. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. E. Rosenman SJ. Preventing suicide: what will worlt and what will not. The Medical Journal of Australia 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate in. Washington, 201?. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 2010:14i1jz24?34. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TRI Swann AC, Powell KE, Potter LB, Kresnow Mj, D'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. Stone DM, Holland KM, Bartholow EN. Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 201?. 12. World Health Organization. Risks to mental health: An overview ofvulnerabilities and risk factors. Geneva: 2012. 13. Caine ED, Reed J, Hindrnan J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo 0. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: StoneASuicide Vital Signs MMWR 3.27.13 [table 1.fig 1} StonegSuicide Vital Signs MMWR 3.27.18 [Tables 2&3) Word Count: 1861f1800 Comment IHJH: Are these references formatted appropriately for The source for ii? is not included: American Foundation for Suicide Prevention. For the location of the is ?Arlington, VA. Comment [Elm]: I'll fix this after! Lreplace #6 once ales weighs in. Rev 3.26.18, e-clearance 2 3 circumstances can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six ii consecutive three-year periods (1999-2016}, were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, DC. Data from the National Violent Death Reporting System, covering 27 states l0 in 2015, were used to examine contributing circumstances among decedents with and without known [6 than decedents with MHP, but were common across groups. [7 Conclusions: Suicide rates increased significantly across most states from 1999?2015. Various circumstances IS contributed to suicides among people with and without known l9 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 20 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 2 after a suicide occurs. 22 INTRODUCTION 23 AND PURPOSE 24 in 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States among people 25 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 2f) urbanization levels Suicide is the 10'h leading cause of death and is one ofjust three leading causes that 2? are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm. a key risk factor 28 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 2?9 nation more than 569 billion in direct medical and Work loss costs 30 The Notional Strategyforsmcide PreventioanSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal}. Such 32 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 33 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 34 mental health problems contributing circumstances include social and economic problems, 35 access to lethal means among people at risk, poor coping and problem- 36 solving skills, and prior suicide attempts i5}. Expanded awareness of the additional circumstances that .32 contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of 3'8 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific 39 trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level 3 40 comprehensive suicide prevention. Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, M59 Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 4 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states esperiencing increases [2 of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, 14 several circumstances, including relationship problems/loss {45.1% vs life stressors [54.2% vs 49.2%} and IS recentfimpending crises (32.9% vs were significantly more likely among those without a known MHP Comment IHJH: Is there a more up?to- date US?based reference than Rosenman 1993? Comment [2:119]: Alex, what about the Zalsman systematic review? if you look at ll all the treatment strategies they appear to outweigh any other strategies and if within the limiting access to lethal means, Ll many of the studies are international. ii 1. Zalsman. G., et al., Suicide ll prevention strategies revisited: matrear systematic review. The Lancet 5! 2015.317}: p. sac-ass. i. ii ==adle::gg= La not.vd-a-Ih-ICIG-Iai-tlnl =Il Cal Rev 326,18, e-clearance 4 METHODS 42 Suicide rates were analyzed for people aged :10 years only, as attributions of suicidal intent in younger children 43 are variable Age?speci?c suicide counts were tabulated based on National Vital Statistics System coded death 44 certi?cate records {international Clossrficotion of Diseases 10"h Revision, underlying-cause?of death codes K60- 45 X34, 1'870, Age-specific population estimates were obtained from US. Census Bureauf'lI-lational Center for 46 Health Statistics bridged~race population data releases. 4? 48 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 49 spanning 1999-2016. Hate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed 50 per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three~year data 5] aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 52 are reported in terms of average annual percentage changes 53 54 Characteristics {Table 2} and circumstances {Table 3} of suicide decedents 210 years, with and without known 55 lleP, were compared in the 2? states with complete data participating in National Violent Death 56 Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and 5? Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance ?38 dependence, which are captured separately in aggregates data from three primary data sources: 59 death certificates, coroner/medical examiner reports {including toxicology}, and law enforcement 60 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression oi analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and 62 race/ ethnicity. 63 RESULTS The most recent overall suicide rates {representing 2014?2016} varied four?fold, from 5.9 to 29.2 65 {Montana} per 100, 000 persons per year {Table Across the study period, rates increased in all states, except or absolute Increases ranging from 8 {Delaware} to +8 1 {Wyoming} per 100 000. Percentage' Increases in rates 68 ranged from +5.91% {Delaware} to {North Dakota}, with increases of more than 30% observed in 25 states 69 {Table 1, Figure 10 TI Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females {43 1?2 states}, as well as for the US. overall {Table Nationally, the model-estimated AAPC for the overall suIcIde 7" rate was By sex, estimated national rate trends further indicated significant increases for males 74 and females TS Suicide decedents without known MHP were compared to those with in 2? states. While all decedents were predominately male {Table 2; 25.8%} and non-Hispanic white those without 1?8 known MHP, relative to those with lleP, were more likely male {33.6% vs. 68.8%; odds ratio 95% CI 80 without known MHP also had significantly greater odds of perpetrating homicide-suicide {adjusted odds ratIo RI 2 El, 95% CI 2.23.8}. Among adult decedents, 20.1% and 15.3% of people without and with MHP, 82 respectively, ever served, or were currently serving, in the U.S. military. While firearms were the most common method of suicide overall and for both groups, deceden'ts 84 without known MHP were more likely to die by firearm {55.3% vs. 40.6%} and less likely to die by 85 {26.9% vs 31.3%} or poisoning {10.4% vs 19.3%} than those without known 36 These differences remained significant in the adjusted models. 2 - [.correct Comment Per Table 2. the range in the loWer bound of the 95% for l} non?white racialfethnic groups is 1.0 {Hispanic} to 1. 5 1 correct to me Jeff was looking at the data a. ~1r - i Ill interpretation. I I ., Comment In?ll: SK, TS, Opinions as to whether this should stay here or be moved to limitations? i?d prefer to keep it here but realize it may not be conventional. Comment Per Table 2, the range in 005 for non?white racial/ethnic groups is 1.2 to 2.0. Please confirm [Comment pane K?r KF Jeff looks to be n. Comment lzn?]: KF, KY, TS?This looks a little differently. For the lower limits here he took the smallest and largest numbers across all LL Us and then did the same for upper limits. l'rn guessing you guys have done this before and it?s accepted? Comment Per Table 2, the range in the upper bound of the 55% for non?white racialfethnic grouos is 1.3 kll-lispanic} to 3.1 {Other}. Please confirm. J, Comment Italy]: KF, KY. Actually this looks like a typo. Should be above comment about Jeff's Rev 3.25.18, e-clearance ft? Decedents without known MHP were less likely to receive toxicology Easting. Among those with toxicology Comment lH-lfl: AW comment on 88 results, decedents without known l'leF were less likely to test positive for any substance overall 95% substance causes of death? 89 but more likely to test positive for alcohol laClFt=1.2, 95% ?TComment T5440 mom .3 99 All suicide decedents with MHP and approximately 85% without known MHP had available [Comment Might be better to report as: 23.0% vs. 40.3% 9 circumstances information {Table People without known MHP were less likely to have any substance abuse problems {309:0}, 95% While two-thirds of those with known MHP had a history of mental health 92 {Comment Might be better to 93 or substance abuse treatment jUst over half were in current treatment. report as: 29.4% vs. toast: Iznf?Sl]: ltF, in, T54 tried to ill preserve aUR?sll Thoughts? I 1 [Comment Add citations to support this statEmentDecedents without known MHP versus those with known MHP had significantly greater likelihood of any 95. relationship problem/loss (45.1% vs. specifically intimate partner problems [30.2% vs. 96 [115% vs. and recently perpetrating interpersonal violence vs. They 9? were also more likely to have experienced any life stressors [54.2% vs 491%], such as criminal?legal problems a Jk I I Comment mm: Alex? add Zalsrnan again? Don?thave room for another ref{109% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 99 the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the i: ll [00 adjusted models. Among those with crises, intimate partner and physical health problems were the most Ell Any comment that the common types for both groups and did not differ between them. ll." ll, 1?99 ranking States are rural and mainli? l" 51: if; the Northwest: MT, AK, WY, UT IUZ Decedents without known MHP had signi?cantly lower odds of recent release from any institution a? Comment Infill: Unfortunately we can?t} :55 accommodate the added text n. 03 95% but among those who were recently released they were significantly more likely to be . l04 released from a correctional facility [253% vs. or hospital vs. 33.0%] than those with a known MHP. Among decedents with known lv'lHP who were recently released from an institution 463% of this il a lilo group were released from facilities. :l ll)? Decedents without known MHP, compared to those with lleP, were significantly less likely to have a history of El lilfi suicidal ideation 95% 95% 0:03-03. Nordthan 1 in i If i .3: [(19 five people in both groups disclosed suicide intent [22.4% vs. ll! i Ill Conclusions and Comments I i From 1999?2015, 44 states saw significant suicide rate increases. Rates increased significantly for males, In 34 i i ?2 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender ?3 gap in male-female suicide rates between half of the states experienced substantial l4 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is [Is necessary. Fortunately, data from the 2? states participating in can shed light on the circumstances that He contributed to recent suicides, and can help guide prevention activities. is common parlance in the field that :suicide is not caused by a 18 single factor;j however, soieideresearch and prevention practices, almost?seleiwlargely focus on identifying and 119 treating current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal~lega matters, I20 eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that rel 22 suggest many suicides and attempts occur with minimal deliberation 3 I i i l23 Among people with MHP. two-thirds had a history of mental health and/or substance abuse treatment and over [24 half were in current treatment. This suggests that additional supports for this population are needed to keep l25 them safe. This includes broader implementation of affordable and effective treatment modalities such as l26 doctor-patient collaborative ca re models and cognitive?behavioral therapy. Additionally, greater access to 127 behavioral health providers inimderserved]areas is needed, as is expansion of healthcare systems to Integrate 3 Ftev 315,18, e-clearance IER physical and behavioral health that better support suicide prevention and patient safety, especially through care [Comment Comment about My? Comment [2:119]: SK, sure 3 there's a weakness in there. I inserted I29 transitions I30 Study findings indicate that people with known leHP also experienced other life stressors such as jobffinancial, relationship andy?or physical health problems. These findings point to the need to both prevent the conditions if language above to state that it ,1 consistently has one ofthe highest rates. 32 associated with mental health problems in the first place, and to support people with known MHP to decrease [33 their risk of poor social, health, and economic outcomes 3 H4 Together, these results underscore the importance of comprehensive statewide suicide prevention activities :5 135 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I l?o economic supports le.g., housing stabilization policies, household financial support}; teaching coping and I problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in [38 life; promoting social connectedness to increase a sense of belongingneSs and access to informational, tangible, I39I emotional, and social support; and identifying and better supporting people at risk. Other strategies include 5 I40 creating protective environments leg, reducing access to lethal means among people at risk, creating i: 4l organizational and workplaCe policies to promote help-seeking, easing transitions into and out of work for 5 I42 people with MHP and other life challenges], supporting family and friends aftera suicide, and assuring safe I43 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning .implement such a comprehensive approach to suicide prevention I45 These findings have at least three limitations. In the state-level analysis, rankings for four states D, MA, RI, I46 might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing I47l reported suicide rates downward], or decreased percentages of such deaths over time {potentially biasing I43 estimated rate trends upward]. is not yet nationally representative; the 27 states included I49 represent 49.6% of the population [50 ISI Third, abstractors of data are limited to information contained in investigative reports. Therefore, the l52 extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth 53 interviews with next?of?kin often see greater attributions to mental disorders however many 15.4 methodological variations across studies exist It is likely that some people without known MHP in the ISS current study were experiencing mental health challenges that were unknown, and hence underreported by key 15s informants. However, any lack of awareness of decedent lv'lHP suggests the importance of addressing the broad range of contributing circumstances. I58 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are 159 available. States and communities can use data from and resources such as CDC's Preventing Suicide: ?50 Technical Package of Policies, Programs, and Practices {11} to better understand their suicide problem and l?l prioritize evidence-based comprehensive suicide prevention. [62 Acknowledgments I63 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital lo4 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. 155 Isa Con?ict of Interest No conflicts of interest were reported. my IoEl Corresponding author: Deborah M. Stone, TED-4386942 I69 I TO Author Affiliations: [7202 203 204 205 206 El}? 208 209 2 I 0 Rev 3.25.18, e-clearance 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis. Research, and Practice Integration, National Center for Injury Prevention and Control, CDC Preference? 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Ctuery and Reporting System i Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Way-Stephenson A2, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001? 2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2015. 4. Kochanek K, Murphy S, In J, Arias E. Mortality in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics 201?. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. E. Rosenman SJ. Preventing suicide: what will worlt and what will not. The Medical Journal of Australia 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate in. Washington, 201?. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 2010:14i1jz24?34. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TRI Swann AC, Powell KE, Potter LB, Kresnow Mj, D'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. Stone DM, Holland KM, Bartholow EN. Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 201?. 12. World Health Organization. Risks to mental health: An overview ofvulnerabilities and risk factors. Geneva: 2012. 13. Caine ED, Reed J, Hindrnan J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo 0. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: StoneASuicide Vital Signs MMWR 3.27.13 [table 1.fig 1} StonegSuicide Vital Signs MMWR 3.27.18 [Tables 2&3) Word Count: 1861f1800 Comment IHJH: Are these references formatted appropriately for The source for ii? is not included: American Foundation for Suicide Prevention. For the location of the is ?Arlington, VA. Comment [Elm]: I'll fix this after! Lreplace #6 once ales weighs in. Rev 326,18, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 2 3- Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, lVlD1 4 Background: Suicide rates in the United StatEs have risen nearly 30% since 1999. Mental health problems 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing a circumstances can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged :10 years, by state and sex, across six ti consecutive three-year periods [1999?2016], were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, Data from the National Violent Death Reporting System, covering 2? l0 states in 2015, were used to examine contributing circumstances among decedents with and without known I lv'lHF. [2 Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases l3 of more than 30%. Rates increased significantly among males and females. in 34 and 43 states, respectively. l4 Dver half of decedents did not have a knowo MHP. Among decedents with circumstance information, IS several circumstances, including relationship problems/loss [45.1% vs life stressors (54.2% vs 49.2%] and I6 rocentfim pending crises [32.9% vs were significantly more likely among those without a known 1? than decedents with MHP. but were common across groups. It: Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without knowo MHP. 20 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 2 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 22 friends/family after a suicide occurs. 23 CTIDN 24 BACKGROUND AND PURPOSE 25 in 2016, nearly 45,000 suicides [age-adjustedli occurred in the United States among peopl 26 .210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 27 urbanization levels Suicide is the 101th leading cause of death and is one ofjust three leading causes that 28 are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm.a key risk factor 29 for suiclde, increased nearly 45% between 2001 and 2015 Together, suicides and self~harm injuries cost the 30 nation more than $69 billion in direct medical and work loss costs 3 The National Strotegyfor Suicide Preventionl'NSSP) calls for a public health approach to suicide prevention 32 with efforts spanning across multiple levels individual, family/relationship, community, and societal]. Such 33 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 34 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 313 access to lethal means le.g., substances, ?rearms, bridges) among people at risk, poor coping and problems solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that 38 contribute to suicide risk apart from and action to address them, can help reach the nation?s goal of 39 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific Comment Is there a more up?to- date US?based reference than Rosenman 1993? Rev 3.26.18, e-clearance 40 trends in suicide rates, assesses the multiple contributing factors, and provida recommendations for multi?level 4 comprehensive suicide prevention. 42 METHODS 43 Suicide rates were analyzed for people aged 3:10 years only, as attributions ofsuicidal intent in younger children 44 are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death 45 certificate records [international Classification of Diseases 10?? Revision, underlying-cause-of death codes X??v 4b X84, Age-specific population estimates were obtained from U.S. Census Bureau/National Center for 4? Health Statistics bridged?race population data releases. 48 49 National and state?level suicide rate estimates were calculated for six consecutive three?year aggregate periods SUI spanning 1999?2016. Rate estimates were age?adjusted to the LLB. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data 5] aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 52 53 are reported in terms of average annual percentage changes 54 55 Characteristics [Table 2) and circumstances {Table 3) ofsuicicle decedents 210 years, with and without known 56 were compared in the 27 states with complete data participating in National Violent Death 5? Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and 58 Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance 59 dependence, which are captured separately in aggregates data from threa primary data sources 60 death certi?cates, coroner/medical examiner reports {including toxicology)' and law enforcement 6 reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression 62 analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and 63 racer'ethnicity. '54 RESULTS 65 The most recent overall suicide rates [representing 2914-2915] varied four-fold, from 5.9 to 29.2 66 {Montana} per persons per year (Table Across the study period, rates increased in all states, except {Wyoming} per 100,000. PerCentage increases in rates ranged from +59% (Delaware) to 67.6% (North Dakota), 69 with increases of more than 39% observed in 25 states {Table 1, Figure TD Tl Modeled suicide rate trends indicated significant increases for 44 states. for males (34 states} and females [43 T2 states), as well as for the U.5. overall {Table Nationally, the model?estimated AAPC for the overall suicide 23 rate was By sex, estimated national rate trends further indicated significant increases for males T4 and females (AAPC ?l5 I vs Suicide decedents in 2? states without known MHP lN=11,039l were compared to those with MHP While all decedents were predominately male {Table 2; 26.8%] and non-Hispanic white without Til known MHP, relative to those with MHP. were more likely male (83.6% vs. 53.8%: odds ratio lDFil=2.3, 95% CI 79 2.2-2.5} and racialfethnic minorities range: Ell-2.1% 95% Cl range Suicide decedents SD without knoWn also had signi?cantly greater odds-of-perpetrating homicide?suicide ladJUsted odds ratio Ell 2.9, 95% CI Among adult decedents, 29.1% and 15.3% of people without and with MHP. 82 33 While ?rearms were the most common method of suicide overall and for both groups, decedents 34 without known MHP were more likely to die by firearm {55.3% vs. 40.5%] and less likely to die by 2 respectively. ever served, or were currently serving, in the US military. {Comment And consistently high rates since 1999?2001. Comment Per Table 2, the range in Cl?s for non-white racialfethnic groups Lis 1.2 to 2.0. Please confirm. .1 Comment Per Table 2, the range in the lower bound of the 95% C15 for non-white racialfethnic groups is 1.0 kli-lispanlc] to 1.6 Comment Per Table 2, the range in the upper bound of the 95% for non?white racialfethnic groups is 1.3 ?lHispanlcl to Please confirm. 4k - JL - Fiev 3.26.18, e-clearance gtlomment Any comment on stran ulation suffocation (25.9% vs 31.3%] or poisoning [10.4% vs 19.3%) than those without known substance causes of death? 85 ban in as MHF. These differences remained significant in the adjusted models. 3? Decedents without known lleP were less likely to receive 88 results. decedents without known l'leP were less likely to test positive for any substance overall 95% E49 but more likely to test positive for alcohol laOFi=1.2, 95% Comment [H.lrj]: Might be better to report as: 23.0% vs. 40.3% Comment Might be better to I report as: 29.4% vs. 103% ,ll Comment Add citations to u' support this statement 90 All suicide decedents with lleP and approximately 85% without known lleP had available 9 circumstances information {Table People without known MHP were less likely to have any substance abuse 92 problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. 94 Decedents without known MHP versus those with known had a?significantly greater likelihood of any 95 relationship problem/loss [45.1% vs. specifically intimate partner problems {30.2% vs. 96 [115% vs. and recently perpetrating interpersonal violence vs. They I 9? were also more likely to have experienced ethee?life stressors {54.2% vs 493%}, such as criminal?legal 98 problems {103% vs. or evictionfloss of home vs. and they were more likely to have had a 99 crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained IUD significant in the adjusted models. Among those with crises, intimate partner and physical health problems Were the most common types for both groups and did not differ between them. It}? Decedents without known MHP had signi?cantly lower odds of recent release Whom any institution 95% 020.4435}, but among those who were recently released they were [04 significantly more likely to be released from a correctional facility (25.7% vs. or hospital {43.7% vs. 33.0%] IOS than those with a known lleP. Among decedents with known MHP who were recently released from an lilo institution 46.7% ofthis group were released from facilities. ll}? Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of liltl suicidal ideation 95% ceasesh and prior suicide attempt 95% More than 1 in fi9 five people in both groups disclosed .. ll'll IIU Conclusions and Comments 111 From 1999?2015, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 I ll states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender I I3 gap in male-female suicide rates between 1999?2014 half ofthe states experienced substantial I I4 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is I I IS necessary. Fortunately, data from the Estates participating in can shed light on the circumstances that 116 contributed to recent suicides, and can help guide prevention activities. II Researchers and practitioners regularly state that suicide is not caused by a single factor;; however, the?feeus?oi I IS suicide research and prevention practices, almost solely, focus on identify and treating ?9 found that more than half of suicide decedents in did not have a known MHP. This group suffered more l20 relationship problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts [22 occur with minimal deliberation [23 Among people with l'leP, two?thirds had a history of mental health andfor substance abuse treatment and over l24 half Were in current treatment. This suggests that additional supports for this population are needed to keep l25 them safe. This includes broader implementation of affordable and effective treatment modalities such as 3 Flev 3.26.18, e-clearance 26 doctor?patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to behavioral health Emderservedhreasjsneeded, as isexpansion_of healthca re systems I28 needed?that?tg integrate physical and behavioral health and patient 29 safety, especially through care transitions I30 Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, I3I relationship, and/or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place, and theneed-to support people with known MHP to [33 decrease their risk of poor social, health, and economic outcomes 134 Toget_he_r,_these the importance of comprehensive statewide suicide prevention I35 activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening I36 economic supports le.g., housing stabilization policies, household ?nancial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems. especially early in I38 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, 39 emotional, and social supportT-Land identifying and better supporting people at risk. Other strategies include I4O creating protective environments leg, reducing access to lethal means among people at risk, creating organizational and Workplace policies to promote help-seeking, easing transitions into and out of work for I42 people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe I43 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning ?l44 and?impiementingto implement such a comprehensive approach to suicide Prevention I45 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, Fil, I46 might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward], or decreased percentages ofsuch deaths over time {potentially biasing I48 estimated rate trends upward). Eeconcg, is not yet nationally representative; the states included I I49 represent 49.6% of the population ISO Third, abstractors of data are limited to information Contained in investigative reports. Therefore, the [51 152 extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth l53 interviews with next-of?kin often see greater attributions to mental disorders however many I54 methodological variations across studies exist It is likely that some people without known lleP in the I55 current study were experiencing mental health challenges that were unknown, and hence by key lib informants. However, any lack of awareness of decedent MHP suggestsJ?even?liurtheef-the importance of I5?r addressing the broad range of contributing circumstances. I58 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are 159 available. States and communities can use data from and resources such as CDC's Preventing Suicide: 0 I60 Technical Package of Policies, Programs, and Practices [11] to better understand their suicide problem and lol prioritize evidence-based comprehensive suicide prevention. 62 Acknowledgments I63 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I64 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. lit-5 I66 Con?ict of Interest No conflicts of interest were reported. I63 Corresponding author: Deborah M. Stone, dstoneB?lcdc?ov Tim-4883942 4 1 1? . Comment IHJU: Any comment that the top ranking states are rural and mainly in the Northwest: MT, AK, WY, UT [Comment Comment about 1159 170 1'11 1'12 1T3 174 ITS lit) 178 179 131] 1111 182 183 134 185 136 18? 138 1119 190 191 1'92 193 194 195 19o 19? 191:1 199 200 201 202 203 204 2115 2116 20? 208 209 2 10 Rev 3.25.18, e-clearance Author Affiliations: 10ivision of Violence Prevention, National Center for Injurv Prevention and Control, 1Division at Anaivsis, Research, and Practice Integration, National Center for lnjurv Prevention and Control, CDC Eleferen cz% 1. Centers for Disease Control and Prevention. Web?based injury Statistics Query and Reporting System 1' Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. 2. Nev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization teveis by Sex, RacefEthnicity, Age Group, and Mechanism of?eath - United States, 2001-2015. MMWR Surveill Summ, 201?. 66118]: p. 1?15. 3. Curtin, 5.12., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1939-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hvattsville, MD. 4. Kochanek, 11., et al., iiviortaiitvr in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 201?. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategvfor suicide prevention: goois and objectives for action: a report of the US Surgeon Generai and of the National Action for Suicide Prevention. 2012. E. Rosen ma n, 5.1., Preventing suicide: what work and what wilt not. The Medical Journal of Australia, 1993. Torguson, K. and A. O'Brien, Leading Suicide Prevention E?orts Unite to Address Rising Nationai Suicide Rate. 201?: Washington, D.C. S. Crepeau?Hobson, F., The Autopsy and Determination of Child Suicides: A Survey of Medicai Examiners. Archives of Suicide Research, 2010. 1411}: p. 24?331. 9. American Association. Diagnostic and statisticai manual of mental disorders {05115601. 2013: American Pub. 10. Simon, T.R., et aI., Characteristics of impulsive suicide attempts and ottempters. Suicide and Life? Threatening Behavior. 2002. 32(51): 49?59. 11. Stone, 0.10., et ai., Preventing suicide: A technicoi package ofpoiicies, programs, and practice. 2012. 12. World Health Organization, Risks to mentai heaith: An overview of and riskfactors. Geneva: WHO, 2012. 13. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practical guidance. lnjurv prevention, 2017: p. injurvprev-2012-04236Ei. 14. Milner, A., J. Sveticic, and D. De Leo, Suicide in the absence ofmentai disorderPA review autopsyr studies across countries. IntJ Soc 2013. sets}: a. 545?54. 15. Pouliot, L. and D. De Leo, Criticai issues in autopsy studies. Suicide Life Threat Behav, 2006. 35(5): p. 491-510. Attachments: Stone_Suicide Vital Signs MMWR 3.27.13 [table 1.ng 1} Stone_Suicide Vital Signs MMWR 3.22.181Tahles 2&3) Word Count: 1861,:? 1800 . Comment Are these references ,r formatted apnropriateiv for The source for it? is not included: American Foundation for Suicide Prevention. For the location of the is Arlington, VA. OJ Mom?s3.27.18 MMWR for e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, (0.0.). Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known MHP. Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs 49.7%) and recentfimpending crises [32.9% vs were significantly more likely among those without a known MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedn occurred in the United States (0.3.), among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racial/ethnic groups, and urbanization levels Suicide is the 101th leading cause of death and is one ofjust three leading causes that are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self?harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strotegyfor Suicide Prevention{NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems (MHP) Other contributing circumstances include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problem- solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific 3.27.18 MMWR for e-clearance trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi-level comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 2:10 years only, as attributions of suicidal intent in younger children are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Classi?cation of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureauf'National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics (Table 2} and circumstances (Table 3) of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and race/ethnicity. RESULTS The most recent overall suicide rates (representing 2014-2016} varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (with the highest current suicide rate}, with absolute increases ranging from +0.8 {Delaware} to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +53% {Delaware} to +57.6% (North Dakota}, with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females (43 states}, as well as for the U.S. overall (Table Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP While all decedents were predominately male {Table 2; 76.8%} and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5} and racial/ethnic minorities (DR range: 122.1; 95% CI range Suicide decedents without known MHP also had significantly greater odds of perpetrating homicidevsuicide (adjusted odds ratio 308 2.9, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. While firearms were the most common method of suicide overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%} and less likely to die by suffocation 2 100I 101 102 103 104 105 106 107 108 109 110I 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 3.27.18 MMWR for e-clearance (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those without known MH P. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems (a0R=0.7, 95% While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP versus those with known MHP, had a greater likelihood of any relationship problem/loss (45.1% vs. specifically intimate partner problems (30.2% vs. arguments/conflicts (12.5% vs. and recently perpetrating interpersonal violence vs. They were also more likely to have experienced other life stressors (54.2% vs such as criminal-legal problems (10.7% vs. or evictionXloss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. Decedents without known MHP had significantly lower odds of recent release 95% CI (0.4-0.5) from any institution, but among those who were recently released they were significantly more likely to be released from a correctional facility (25.2% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MH P, were less likely to have a history of suicidal ideation 95% and prior suicide attempt 95% More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male-female suicide rates between 1999-2014 Overall, half of the states experienced substantial increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is necessary. Fortunately, data from the states participating in can shed light on the circumstances that contributed to recent suicides, and can help guide prevention activities. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MHP, two~thirds had a history of mental health and/or substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as 3 126 127 128 129 130 131 132 133 134 135 136 137 133 139 140 141 142 143 144 145 146 147 148 149 1513 151 152 153 154 155 156 157 I58 159 160 161 162 .163 164 .165 166 167' 168 3.27.13 MMWR for e-clearance doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as jobg'financial, relationship, andfor physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help?seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to Prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state?level analysis, rankings for four states D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in?depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence unreported by key informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as CDC's Preventing Suicide: a Technical' Package of Policies, Programs, and Practices [11] to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488-3942 4 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 3.27.18 MMWR for e-clearance Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atlanta, GA: National Center for injury Prevention and Control. Retrieved March 15, 20.18. 2016. 2. Ivey-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66118}: p. 1-16. 3. Curtin, S.C., M. Warner, and H. Hedegaard, increase in suicide in the United States, 1999-2014. 2016, US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, MD. 4. Kochanek, K., et al., Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Rosenman, 5.1., Preventing suicide: what will work and what will not. The Medical Journal of Australia, 1998. 16912}: p. 100402. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. 2017: Washington, D.C. 8. CrepeauwHobson, F., The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research, 2010. 14(1}: p. 2434. 9. American Association, Diagnostic and statistical manual of mental disorders P). 2013: American Pub. 10. Simon, T.R., et al., Characteristics of impulsive suicide attempts and attempters. Suicide and Life? Threatening Behavior, 2002. 32(51}: 49?59. 11. Stone, D.M., et al., Preventing suicide: A technical package afpoiicies, programs, and practice. 2017. 12. World Health Organization, Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO, 2012. 13. Caine, E.D., et aI., Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention, 2017: p. injuryprevu2017?042366. 14. Milner, A., J. Sveticic, and D. De Leo, Suicide in the absence of mental disorder? A review autopsy studies across countries. Int Soc 2013. 59(6): p. 545?54. 15. Pouliot, L. and D. De Leo, Critical issues in autopsy studies. Suicide Life Threat Behav, 2006. 36(5): p. 491610. Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 [table 1.fig 1} Stone_Suicide Vital Signs MMWR 3.27.18 (Tables 2&3] Word Count: 1861,? 1800 Rev 3.23.18, e-clearance Short Title: Vital Signs: Contributing CircumstancEs to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Ken-ring Yuan, 3 Kristin M. Holland, PhD;1Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 4 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems 5 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing a circumstances can inform comprehensive state suicide prevention planning. 7 Methods: Trends in age-adjusted suicide rates among people aged 210 years, by state and sex, across six fl consecutive three-year periods [1999-2016], were assessed using data from the National Vital Statistics System 9 for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 27 states Ill in 2015, were used to examine contributing circumstances among decedents with and without known MHP. I I Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases [2 of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. @1100]?! decedents in 2015 in 2? states, nger half154.0%] of decedents did not have a known MHP. Among 14- decedents with circumstance information, several circumstances, including relationship problems/loss {45.1% vs I5 life stressors [54.2% vs 49.2%) and recent/impending crises {32.9% vs were significantly more [6 likely among those without a known MHP than decedents with MHP, but were common across groups. I7 Conclusions: Suicide rates increased significantly across most states from 1999?2015. Various circumstances I8 contributed to suicides among people with and without known [9 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 20 to revent suicide risk before it occurdtidentify and support people at risk, prevent reattempts, and help 2 after a suicide occurs. 22 INTRODUCTION 23 BACKGROUND AND PURPOSE 24 in 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States among people 25 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 2f) urbanization levels Suicide is the 10th leading cause of death and is one ofjust three leading causes that 2? are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm. a key risk factor 28 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 29 nation more than 569 billion in direct medical and work loss costs 30 The National StrotegyforSuicide PraventioanSSP} calls for a public health approach to suicide prevention 3 with efforts spanning across multiple levels individual, family/relationship, community, and societal). Such 32 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 33 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 34 mental health problems or providing follow-up care to people who have attempted suicide Other 35 contributing circumstances include social and economic problems, access to lethal means substances, 36 firearms. bridges} among people at risk, poor coping and problem?solving skills, and prior suicide attempts l5). Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action 38 to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in 39 achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing 40 factors, and provides recommendations for multi-level comprehensive suicide prevention. 1 ., Comment I think some sort of clarification is needed to distinguish from previous two sentences since what is reported here are not data on full examined time period or all 50 states. Tracked an idea but there are likely other Lways to clarify. rComment Inig?l: Suggest change in wording here for flow: In 2015, more than half of decedents fin 2? states did not {just suggestion not required Lchange] Comment Is it prevent suicide RISK before it occurs? Or prevent suicide? ?l-laven?t seen it phrased that Way before. Fiev 3.23.18, e-clearance Comment Inigll]: This is unclear to me. DO YOU mean that assigning l?tE?t in a 4 METHODS 42 SUICICIE- rates were analyzed for?p-eople aged :10 years only, asiattributlons of suicidal intent in younger children younger person is variable or difficult to 43 are determine? Not sure variable is the 44 certificate records {international Classification of Diseases 10"? Revision, underlying-cause?of death codes XED- Lmryect word 1 i? Comment Suggest pulling out the references to tables 2 and 3 since table 1 ti hasn?t been mentioned yet and helps saves on word count. if you want to keep i them in, suggest adding ?table 1" ksomewhere into the above paragraph. 1 45 X34, YBTD, Age-specific population estimates were obtained from U.S. Census Bureau/National Center for 4s Health Statistics bridged?race population data releases. 4? 48 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 49 spanning 1999-2016. Hate estimates were age-adjusted to the LLS. year 2000 standard population and expressed 50 per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three~year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends i 1- Comment Given attention to this particular at-risk population in the results, SI 52 are reported in terms of average annual percentage changes 53 i made me wonder if some mention needs of this group should be in the discussion. There are other higher risk groups described below. A consideration 55 MHF, were compared in the 2? states with complete data participating in National Violent Death 56 Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and 5? Statistical Manual of Mental Disorders, Fifth Edition with the Exception of alcohol and other substance 58 dependence, which are captured separately in aggregates data from three primary data sources 59 death certificates, coroner/medical examiner reports {including toxicology}, and law enforcement 60 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and I i could be in the discussion about i prevention needs of underserved areas {bottom of page 3] some phrasing could i be added about the need for services for i higher risk groups. such as those who 5' have served In the militarv and those who i have difficulties, such as MPH, Intimate partner problems, have physical health problems, and have financial and legal dif?culties. Or on page 4 where 1 "identifying and better supporting people Lat risk? is stated an e.g. could be added. 62 race/ ethnicity. 63 RESULTS as The most recant overall suicide rates {representing 2014?2016} varied four?fold, from 5.9 to 29.2 to {Montana} per 100,000 persons per year {Table Across the study period, rates increased in all states, except as Nevada {which had a consistently high rate throughout}, with absolute increases ranging from +0.3 {Delaware} or to +8.1 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% {Delaware} to {North 68 Dakota}, with increases of more than 30% observed in 25 states [Table 1, Figure Comment It appears non-M HP is being compared to non-MHF. Tracked what i think the edit is but please double kcheck. 69 "i0 Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females {43 TI states}, as well as for the US. overall {Table Nationally, the mode ~estirnated AAPC for the overall suicide 7?2 rate was By sex, estimated national rate trends further indicated significant increases for males T3 and females "i4 i5 Suicide decedents without known MHP {N=11,035i} were compared to those with MHP in 2? states. To While decedents were predominately male {Table 2; 76.8%} and non-Hispanic white those without ii known MHP, relative to those with MHF, were more likely male {33.5% vs. 53.3%; odds ratio 95% CI i8 2.2?2.5} and racialiethnic minorities range: Suicide decedents without known MHF also had i 1? i9 significantly greater odds of perpetrating homicide?suicide {adjusted odds ratio 2.9, 95% CI 80 Among adult decedents, 20.1% and 15.3% of people without and with MHF, respectively, ever served, or were Il i l' i til currently serving, in the U.S. milita I 82 While use of firearms wen?I14 we;- the most-common method of suicide 83 decedents without known MHP were more likely to die by firearm {55.3% vs. 40.5%} and less likely to die by I fir-l- {26.9% vs 31.3%} or poisoning {10.4% vs 19.3%} than those without known i i 35 MH1These differences remained significant in the adjusted models. 2 Fiev 3.23.18, e-clearance 86 Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology 8? results, decedents without known l'leP were less likely to test positive for any substance overall 95% 88 but more likely to test positive for alcohol iaClFt=1.2, 95% 89 All suicide decedents with lv?lHP and approximately 85% without known lv'lHF' had available 90 circumstances information {Table 31. People without known lv'lHP were less likely to have any substance abuse 9 problems 95% While two-thirds of those with known MHP had a history of mental health 92 or substance abuse treatment jUSt over half were in current treatment. 93 Decedents without known MHP versus those with known MHP had significantly greater likelihood of any 94 relationship problemfloss (45.1% vs. specifically intimate partner problems {30.2% vs. 95 arguments/conflicts [123.5% vs. and recently perpetrating interpersonal violence vs. They I so aigowere also more likely to have experiencad any life stressors (54.2% vs Such as criminal?legal 9? problems vs. or evictionfloss of home vs. and they were more likely to have had a 98 crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained 99 significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. IUI Decedents without known MHP had signi?cantly lower odds of recent release from any institution IDZ 95% but among those who were recently released they were significantly more likely to be I'll}: released from a correctional facility [251% vs. or hospital l43.?% vs. 33.0%] than those with a known il4 MHP. Among decedents with known lv'lHP who were recently released from an institution 453% of this IDS group were released from facilities. lilo Decedents without known MHP, compared to those with lleP, were significantly less likely to have a history of If}? suicidal ideation [23.0% vs. 40.3%] and prior suicide attempts l1?.3% vs. More than 1 in five people in [08 both groups disclosed suicide intent [22.4% vs. lil9 Conclusions and Comments From 1999?2015, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender 12 gap in male-female suicide rates between 1999?2014 Overall, half of the states experienced substantial 13 increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is I IS contributed to recent suicides, and can help guide prevention activities. If) Eiesearchers and practitioners regularly state that 'suicida is not caused by a single factori however, research and prevention practices often focus on identifying and treating lv'lHP The current study found that more llti than half of suicide decedents in did not have a known l'v'lHP. This group suffered more relationship I I9 problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. 120 This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal l2l deliberation ill] [23 half were in current treatment. This suggests that additional supports for this population are needed to keep IE4 them safe. This includes broader implementation of affordable and effective treatment modalities such as l25 doctor-patient collaborative care models and cognitive?behavioral therapy. Piklditionaliy, greater access to 126 behavioral health providers in underserved areas is needed, as is expansion of healthcare systems to integrate 3 suggests at least to me there is limited ., Comment Inigil]: This statement needs a reference. Not sure if reference 6 covers Comment [Fill]: True point but it prevention opportunity. Given 22% disclosed suicide intent, perhaps there needs to be more stated here. Idea: suicides can occur with minimal deliberation but many individuals disclose suicidal intent or demonstrate other risk factors that help in ideritifying at-risk individual who need immediate access to ,?pportive services. .alh Comment Wonder if it might be good to move this point down into the paragraph about comprehensive strategies or its own paragraph since it is applicable to those with known and not known MPH. if moved out of this paragraph specific to MPH, and framed more broadly about what all persons need, then attention to a cauple of vulnerable populations per earlier ll4 necessary. Fortunately, data from the 2? states participating in can shed light on the circumstances that l22 liimong people with MHP, twovthirds had a history of mental health andfor substance abuse treatment and over comment could be added to this idea. I28 I29 3fl I31 I32 [34 I35 I36 13? I33 I40 I4I I42 I43 I44 I45 I46 I43 I4?:ll I50 I52 153 I54 155 I56 158 I59 I?l I62 I63 I64 I135 166 lb? I158 I69 Rev 3.23.18, e-clearance physical and behavioral health that better support suicide prevention and patient safety, especially through care transitions Study findings indicate that people with known MHP also experienced other life stressors such as joblfinancial, health problems. These associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor social, health, and economic outcomes Together, these results underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports leg, housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support: and identifying and better supporting people at risk. Other strategies include creating protective environments leg, reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined Intent [potentially biasing reported suicide rates downward], or decreased percentages of such deaths over time [potentially biasing estimated rate trends upward]. Second, is not yet nationally representative; the 2? states included represent 49.6% thhe population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next?of?kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent lv'lHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can Use data from and resources such as CDC's Preventing Suicide: Technical Package of Policies, Programs, and Practices {11} to better understand theirsuicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Con?ict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, Author Affiliations: Comment IFCII: I like the evolution I have seen of the results and discussion to emphasize the dif?culties of those without known MHP. I do think it?s important as done for the discussion to talk speci?cally about those with known MPH. But, the two MPH group paragraphs in the dlscussion seem to tip the balance more toward the MPH group. I suggest weaving into the previous paragraph that those with known had other dif?culties rather than this being a standalone paragraph. Aiso consider If the prevention implication described here is also applicable to those without MPH. There seemed to be a gap in discussion paragraph #2 where similar precipitating circumstances are noted for the non-M HP group but no prevention strategies are presented. Prevantion strategies for these types of risk factor could be described either in the non?MPH paragraph or this paragraph cold be framed more broadly about the prevention needs of both non? LMPH and i'leH groups. 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.28.18, e-clearance 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. A. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et aI. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives ofSuicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, rroII PW. Characteristics of impulsive suicide attempts and attempters. Suicide and LifeAThreatening Behavior 11. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 12. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev~2017~042366. 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1] Stone_Suicide Vital Signs MMWR 3.27.18 (Tables 28:3} Word Count: 185871800 LnRev 3.23.18, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods {1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National ?v?iolent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known MHP. Results: From 1999?2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, several circumstances, including relationship problemsfloss {45.1% vs life stressors {54.2% vs 49.7%) and recent/impending crises {32.9% vs were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999?2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjusted? occurred in the United States (US), among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10?1 leading cause of death and is one ofjust three leading causes that are increasing 4). Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self-ha rm injuries cost the nation more than $69 billion in direct medical and work loss costs The Notionoi Strotegyfor Suicide Prevention(NSSP,i (5) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi?determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems or providing follow-up care to people who have attempted suicide Other contributing circumstances include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problem-solving skills, and prior suicide attempts (5). Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action to address them, can help reach the nation's goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi-Ievel comprehensive suicide prevention. 1 Rev 3.23.18, e-clearance METHODS Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in younger children are variable (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Classification of Diseases 10?r Revision, underlying-cause-of death codes X60- X84, Y87.0, U03). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridged-race population data releases. National and state?level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCS). Characteristics (Table 2) and circumstances (Table 3) of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception ofalcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and race/ethnicity. RESULTS The most recent overall suicide rates (representing 2014-2016) varied four-fold, from 6.9 (D.C.) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +59% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states) and females (43 states], as well as for the U.S. overall (Table 1). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%) and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5) and racial/ethnic minorities (OR range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%) and less likely to die by (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those without known MHP. These differences remained significant in the adjusted models. 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120I 121 122 123 124 125 126 Rev 3.28.18, e-clearance Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP (1429,40?) and approximately 85% without known MHP had available circumstances information (Table 3). People without known lleP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP versus those with known MHP had significantly greater likelihood of any relationship problemfloss {45.1% vs. specifically intimate partner problems {30.2% vs. arguments/conflicts (12.5% vs. and recently perpetrating interpersonal violence vs. They were also more likely to have experienced any life stressors {54.2% vs such as criminal?legal problems (10.7% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. Decedents without known MHP had significantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be released from a correctional facility {25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.2% of this group were released from facilities. Decedents without known MH P, compared to those with MH P, were significantly less likely to have a history of suicidal ideation {23.0% vs. 40.8%) and prior suicide attempts (10.3% vs. More than 1 in five pepple in both groups disclosed suicide intent {22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant suicide rate increases. Rates increased significantly for males, in 34 states, and for females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male-female suicide rates between 1999-2014 Overall, half of the states experienced substantial increases in suicide rates of more than 30%. Additional research into the specific causes of these trends is necessary. Fortunately, data from the 27 states participating in can shed light on the circumstances that contributed to recent suicides, and can help guide prevention activities. Researchers and practitioners regularly state that 'suicide is not caused by a single factor;? however, research and prevention practices often focus on identifying and treating MHP (6). The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MHP, two-thirds had a history of mental health and/or substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems to integrate 3 Rev 3.28.18, e-clearance physical and behavioral health that better support suicide prevention and patient safety, especially through care transitions (11). Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, relationship, andi'or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor social, health, and economic outcomes Together, these results underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support}; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk. Other strategies include Creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time [potentially biasing estimated rate trends Upward}. Second, is not yet nationally representative; the 27 states included represent 49.5% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: Technicoi Package of Poiicies, Programs, and Practices (11) to better understand their suicide problem and prioritize evidence-based comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gg 770438?3942 Author Affiliations: 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.28.18, e-clearance 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. A. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et aI. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives ofSuicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, rroII PW. Characteristics of impulsive suicide attempts and attempters. Suicide and LifeAThreatening Behavior 11. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 12. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev~2017~042366. 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. IntJ Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1] Stone_Suicide Vital Signs MMWR 3.27.18 (Tables 28:3} Word Count: 185871800 Table 1. Trends in Suicide Rates among Persons 1: 19 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 109,999 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2091 2002 2904 2095 2007 2903 2010 2011 2913 2914 2015 Rank (State Rank) 11 (State Ragnk) H. Both 12.3 (We) 1271+ 0.4) 12.9 0.2) 13.31+ 0.9) 14.5 1+ 0.3) 1541+ 0.9) 1.5 c111154.01) nia 3.1 (Na) 25.4 a. (nia) U.S. Male 29.9 (nia) 2121+ 0.4) 2131+ 0.0) 22.5 23.51+1.0) 24.51+1.0) Female 4.7 5.0 1+ 0.3) 5.3 1+ 0.2) 5.7 1+ 0.4) 5.2 1+ 0.5) 5.9 1+ 0.7) 2.5 %1p<.01) Both 14.3 (nia) 13.41- 0.9) 14.1 1+ 0.5) 15.51+1.5) 15.41+ 0.7) 17. 51+ .1) +1.5 %1p<.05) 25 3.1131) 21.9 %133) AL Male 25.1 (nia) 23.4 (- 1.7) 24.41+ 1.0) 25.41+ 2.0) 27.5 1+ 1.1) 29.1 1+ 1. 5) 1. 3 ?/61 (cs. 05) Female 5.1 (nia) 4.31- 0.3) 501+ 0.2) 1 5.41+ 0.3) .01+ 0. 7) 2. 5 5'41 01) Both 21.0 (nia) 2431+ 3.3) 24.21- 0.5) 2501+ 1.7) 25.41- 0.5) 2331+ 3.4) +1 7 (psi 05) 2 7.31 4) 37.4 113) AK Male 33.2 (nia) 3311+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 0.1) 42. 9 1+ 2. 3) +1.4 1p<. 01) Female 3.51nia) 11.41+ 2.9) 9.31-1.5) 11.1 1+ 1.2) 991-12nis Both 17.31nia) 1351+ 0.7) 19.1 1+ 0.5) 19.11- 0.0) 20. 4 1+ 1.3) 20. 9 1+ 0. 5) +1.0 %11.p< 01) 15 3.1132) 17.3 142) .42 Male 29.3 (nia) 3021+ 1.0) 30. 5 1+ 0.4) 30. 2 1? 0.5) 32. 0 1+ 1.1pc. 05) Female 7.1 (nia) 7.51+ 0.4) 21+ 0.7) 51+ 0.5) 21+ 0.5) 91+ 0. 5) 2. 2 1p<. 01) Both 15.5 We) 1531+ 0.3) 15.21+0.5) 1751+ 1.4) 1921+ 1.5) 21 g..21+20) 01) 12 +5.7114) +35.3%115) AR Male 25 71nia) 2571+ 0.0) 27. 2 1+ 0.5) 23. 2 1+ 1.0) 31. 7 1+ 3.5) 33. 5 1+ 1. 9) 1.5 5'41 1pc. 05) Female 5.5 (nia) 5.91+ 0.3) 5. 2 1+ 0 4) 91+ 1.7) 7.51- 0.4) 51+ 2.1) 3. 5 31:1 01) Both 10.5 (nia) 11.3 11 .01- 03) 12.01+1.0) 11.31- 0.1) 121 05) 45 1.5145) +14.3%145) CA Male 17.9 (nia) 1341+ 0.5) 17.7-( l0.7) 19.1 1+ 1.4) 13.9 (-0.2) 19.21+0.3) +05% nis Female 4.1 (nia) 501+ 09) 4. 0. 1) 41+ 0.5) 5. 3 1- 0.1) .51+ 0. 3) 7 5'41 1pc. 05) Both 17.31nia) 19.21+1.9) 19. 01-0 2) 20. 0 1+ 1.0) 21.51+1.5) 2321+ 15) 1.3 (psi 01) 3 5.9112) 34.1 122) CO Male 23.5 (nia) 3091+ 2.3) 30.510 .4) 31.51+1.0) 33.41+1.9) 35. 3 1+ 2. 9) +1.4 1p<. 01) Female 7.0 (nia) 3.21+1.3) 321+ 0.0) 11+ 0.9) 10.1 1+ 1.?/61 (cs. 01) Both 9.5 (nia) 3.91? 0.7) 9.1 1+ 0.2) 10. 2 1+ 1.1) 11.01+ 0.3) 1151+ 0.5) +1.5 (cs. 05) 45 1.9143) +192 11211134) CT Male 15.4 (nia) 14.5 1? 1.3) 1501+ 0.4) 15. 5 1+ 1.5) 17. 5 1+ 1.0) 17.31- 0.3) 0.9 his Female 3.5 (nia) 331+ 0.2) 3.71- 0.2) 41+ 0.7) 91+ 0.5) 5.21+ 1.3) 3.5 %1p<.05) Rates are age-adjusted to the US. year 2000 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest percentage increase 11} to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 20 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank} 1? (State REnk} Both 13.6 1n1a1 12.21.41 1191? 0.31 13.61+ 1.71 14.21+ 0.61 14.41+ 0.21 0.9 nis 42 0.81501 5.9 ?it: 1501 DE Male 23. 0 1n1a1 20.31.71 1991- 0.41 23.1 1+ 3.21 22. 71- 0. 41 2351+ 0.81 0.6 ?fa Female 31ni'a1 010 .21 4.61? 0.41 91+ 0.31 6. 41 1.51 21- 0.21 1.6 nl's Both 91nia1 6.41+ 0.51 6.41? 0.01 31+ 0.81 6?01.71 6.91+ 0.31 0.9 ?fa 51 1.01481 16.1 ?it 1451 DC Male 10.71n1a1 11.11+ 0.41 10.31; 0.81 12.71+ 2.41 10. 012 .61 11. 71+ 1.71 0.3 ?it: nis Female 1.71nia1?FT 2.31+ 0.611?r 331+ 1.01 61- 0.71 3.61+ 1018-0181 3.5 ?fa Both 14.81n1a1 1521+ 0.41 14.91- 0.31 16. 3 1+ 1.41 16. 31?0 .1 16.41+ 0.11 0.8 301p<.051 29 1.61451 10.6 i131481 FL Male 24.3 1n1a1 24.41+ 0.11 23.61- 0.81 26. 2 1+ 2.61 25. 61 0. 61 25.6 1?0. 11 0.5 ?it. Ms Female 6 3 (ma) 681+ 0.51 6.81+ 0.0111+ 0.31 7.61+ 0.51 81+ 0.31 1.4 301p<011 Both 12.91n1a1 13.21+ 0.31 12.31? 0.91 13. 2 1+ 0.91 13.71+ 0.51 1501+ 1.31 0.9 ?it; nits 39 2.1 1401 +162 131441 GA Male 22.1 1n1a1 23.1 1+ 1.01 21311.81 .91+ 0.61 2261+ 0.71 24.-41+ 1.71 0.5 ?fa nl's Female 5.0 1n1a1810.21 4.6 1- 0.2151+ 0.91 5.8 1+ 0.31 6.6 1+ 0.81 2.1 ?fa 1p<.051 Both 12.91nla1 11. 1 11.81 10.31- 0.71 14. 5 1+ 4.11 14.41? 0.11 1521+ 0.81 2.0 ?fa 35 2.41351 18.3 I1111381 HI Male 20.4 1n1a1 17.21.11 15.3 1- 1.91 21. 91+ 6.71 2251+ 0.51 24.31+ 1.81 2.1 ?fa Female 5.4 1n1a1010.41 551+ 0.5111+ 1.51 21- 0 91 5.91- 0.31 1.2 ?fa nl's Both 17.31n1a1 1921+ 2.01 18.31? 0.91 21. 61+ 3.31 2191+ 0.31 24.71+ 2.81 2.3 3611:1101} 6 7.51 61 43.2 1 71 ID Male 28.4 1n1a1 33.1 1+ 4.71 31.1 1? 2.01 3 91+ 3.81 71- 0 21 3801+ 3.31 1.6 301p<051 Female 7.2 1n1a1 6.1 1- 1.11 6.1 1+ 0.0101+ 2.91 9.51+ 0.51 11.81+ 2.31 4.4 1p<.051 Both 9.9 1n1a1 9.81? 0.11 9.71- 0.11 10. 6 1+ 0.81 11.21+ 0.61 1221+ 1.01 1.5 3131p<.051 44 2.31381 22.8 ?it 1321 IL Male 17.1 1n1a1 16.71? 0.41 16.2 1- 0.41 17. 61+ 1.41 18.51+ 0.91 1981+ 1.31 1.1 36113105) Female 3.7 1n1a1 3.814 0.01 381+ 0.2121+ 0.41 4.51+ 0.41 521+ 0.61 2.4 301p<011 Both 13.01n1a1 13.71+ 0.71 14.41+ 0. 71 1491+ 0.51 16.41+1.41 17.11+ 0.71 +1.9 31: 1p<.011 26 4.11231 31.9 13125} IN Male 22.4 1n1a1 23.21+ 0.81 24. 41+ 1.21 24. 71+ 0.41 2671+ 2.01 28.31+ 1.61 1.5 36113101} Female 4.6 1n1a1 501+ 0.41 531+ 0.2191+ 0.61 681+ 0.91 6.61? 0.21 2.7 361131011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia1 is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 (-1.4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nia) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 i1110.14.01) 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nia) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nia) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia.4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 ?it; 27 3.8 29.3 14(26) LA Male 22.9 (nia) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa Female 4.8 (nia) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 31;? Both 14.5 (nla) 1310) 18.9 3.5) 18. 0.4) 2.2 51: 21 4.0 (25) 2?.4 ?11} (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nia) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nia) 10.3 0.3) 10.1 0. 2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(02) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nia) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 90(4 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (nia) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nia) 4.8 0.0.9) 2.8 31;? Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State ank) Both 10.71n1a) 11.51+ 0.9) 12.41+ 0.5) 12.91+ 0.5) 14.21+1.3) 15. 01 0. 9) 2.3 701p<.01) 35 4.3119) 40.5 fit-r1 5) MN Male 15.3 1nia) 1931+ 1.1) 20.41+ 1.0) 2091+ 0.5) 22.9 81+ 1.9) 2331+ 0.4) 1.7 ?fa 1p<.01) Female 3.5 1n1a) 4.21+ 0.5) 51+ 0.5) 1 1+ 0.4) 05) 591+ 1.2) 4.2 55113101) Both 12.91nia) 14.1 1+ 1.2) 14. 7 1+ 0. 5) 1551+ 0.5) 1551+ 0.1) 15. 210 .3) +1.1 3511:1105) 35 2.3135) 17.8 i55140) M5 Male 22.9 1n1a) 24.5 1+ 1.7) 25.1 1+ 0 5) 2551+ 1.7) 25.910 .9) 25. 31.5) 0.7 nlis Female 4.3 1nia) 501+ 0.541+ 0.9) 210.2) 2.4 1p<.01) Both 14.71n1a) 14.1 1? 0 5) 15.41+1.3) 1501+ 0. 7) 17.51+ 1.7) 20. 0 1+ 2.3) 2.2 15 5.3115) 35.4 i15117) MD Male 25.3 1nia) 23.71? 1.5) 25. 5 1+ 1 .9) 2551+ 1.0) 2591+ 2.3) 32. 21+ 3.3) 1.5 501p<.05) Female 5 41n1a) 5.41+ 0.1) 11+ 0. 7) 31+ 0.2) 7.41+ 1.1) 51+ 1.2) 3.2 501p<.01) Both 21.1 1nia) 2251+ 1.4) 23. 5 1+ 1 .0) 24.71+1.1) 2571+ 2.0) 29.21+ 2.5) 2.1 1 5.01 2) 35.0 35111) MT Male 35.9 1n7a) 3731+ 0.0.1) 41 0.1+ 1 .4) 4551+ 4.4) 1.3 55113101) Female 5.71nia) 5.41+1.5) 41 0.1) 10. 0 1+ 1. 5) 12.51+ 2.5) 1311+ 0.5) 4.5 3511:1101) Both 12.71ni'a) 1221? 0 5) 12. 51+ 0 .4) 11. 71?0 5) 1351+ 1.5) 14.51+1.3) +1.0 ?fa 40 2.1 142) 15.2 i111143) NE Male 22.21nia) 2071-1523.91+19.) 0.5 ?fa nis Female 3.5 1n7a) 4.21+ 0.4)11+ 0 9) 01 1.2.5 ?70 his Both 23.31nia) 22.51- 0 5) 22.1 1 0. 5) 2251+ 0. 5) 21.4 2) 23.1 0.2 ?it. n15 9 0.2151) -1.0 55151) NV Male 35.3 1n1a) 35.714 1.7) 35.1 1 1 .5) 35510.7 51; 1115 Female 5.91nia) 951+ 0.5) 51+ 0.1) 100111.21+ 0. 5) +1.5 ?fa 1p<.01) Both 13.51n1a) 12.51?1.) 13.31+0.5) 15.21 1. 9) 15.51+0.5) 20.0 +2.7 3131p<.05) 17 +5.51 5) +453 %1 3) NH Male 22.51n1a) 21.1 11.4) 21.71+ 0. 5) 24. 5 1+ 3.2.2 5511:1105) Female 5.3 1n1a) 510.5) 91+ 1.3.9 501p<.05) Both 7.5 1n1a)7?01 1) 51 0.2 01+ 0.551.5147) 19.2 35135) NJ Male 13.01n1a) 13.11+ 0.0) 12.51 0.55) 13. 71+ 1.1 14.51+ 0.5) 1451+ 0.1) 0.9 %1p<.05) Female 3.2 1nia) 91? 0.3) 301+ 0.0) 91 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2006 2010 2011 - 2013 2014 2016 Rank 5 {State Rank) 17 (State REnk} 55m 22.01n1a1 2201- 011510.21 2301+ 1.21 24.1 1+ 1.11 25.01+1.91 +1.1351p<.051 4 4.01241 15.3 351391 NM Male 36.6 1nfa1 3737.1 1+ 1 .31 4071+ 3.61 0.4 35 nis Female 5.5 1n1a1411.1121011.71+ 0. 91 1201+ 0.31 3.3 351114.051 Both 7.2 1n1a1 110 .1+1. 11 310 .11 21351114011 49 2.1 1411 25.5 351271 NY Male 12.51n1a1 12.210.31 1291+ 0.71 1391+ 1. 01 1541+ 1.41 14.510 .51 1.4 351p<.051 Female 2.71nfa161-0.11 301+ 0311+ 0. 51 4.2 1+0. 71 461+ 0.51 +4.2 35 1p<.011 55m 13.61nl'a1 13510 .11 1371+ 0.11 14.21+0.51 14.5.1+041 15..31+051 +0.5 35154011 34 1.71441 12.7 35 1471 no Male 22.7 1n1a1 2271+ 0.01 22.2 1- 0.51 2331+ 1.11 2331n75 Female 5 61nfa1 51- 0 21 621+ 0.2.0 35113905} Both 13.31nl'a1 14.61+1.31 1601+ 1.41 16. 6 1+ 0. 61 1641+ 1.91 20. 9 1+ 2. 51 2.9 351p<.011 14 7.61 51 57.6 351 11 no Male 21.4 1n1a1 2451+ 3.21 23012.5 351114.011 Female 5.5 1n1a1 51- 1.01 371- 0.5171+ 2 01 5. 7 1+ 1.01 551+ 1 5.1 3.9 35 n75 Both 11.5 1n1a1 1231+ 0.51 131 1+ 0.51 1341+ 0.21 14.51+ 1.41 15.51+1.01 +2.0 35 1p<.011 32 +4.21211 +350 351191 OH Male 20.4 1n1a1 2091+ 0.51 2221+ 1.31 221p<.011 Female 4.0 1n1a1 471+ 0.71 491+ 0.1131+ 0.51091 5. 7 1+ 0. 51 3.4 351114.011 Both 17.0 1n1'a1 16. 511641+ 1.11 20. 7 1+ 2 31 2351+ 2.61 2.3 35109051 7 6.41101 37.6 i151121 OK Male 26.5 1nfa1 27.31.21 2761+ 0. 51 3031+ 2.51 3341+ 3. 11 3731+ 3.61 2.0 351139051 Female 6.6 1nfa1401- .21 7. 5 1+ 1.11 01- 0.51 651+ 1.61 10.31+ 1.61 2.9 1p<.051 Both 16.41nta1 17.71+1.31 17.71.01 16. 61+ 0.91 19. 6 1+ 1 .21 21.1 {+1.31 +1.6 351p<.011 13 4.61161 26.2 i551261 OR Male 27.41nl'a1 2951+ 2.11 25.51 0. 91 29.51+1.01 31.41+1.51 33 01+ 1.51 +1.1 351154.011 Female 6.51nfa1 1 1+ 0.61 7.71+ 0.6141+ 0.71 6612.7 35109011 Both 12.1 We] 1251+ 0.41 12.61+ 0.31 13. 9 1+ 1.11 15. 0 1+1. 11 16.31+1.21 2.0 351p<.011 30 4.1 1221 34.3 351211 PA Male 21. cm n1a a1 2131+ 0.31 21 .91+051 231 1+ 1.21 2471+ 17.1 25.11+1.31 1.5 351114.011 Female 4.2 1nl'a1 451+ 0.31 451+ 0.0141+ 0.91 5. 0 1+ 0.5111+ 1. 11 3.5 35 1114.011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period1" Current Overail Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 17 (State REnk} Both 9.41n1a10103101 0.0 01 1281+ 3. 81 11.91.91 12. 6 1+ 0. 71 2.6 ?fa 1p<.051 43 3.2130 W1 34.1 ?fa 123 W1 RI Male 15.41n7a1 1521-0 .21 14.81- 0. 31 21.21+6.41 1912?: 2.01 19..61+041 +2.2 ?fa Female 4.0 1n1a1310.7181+ 0. 41 11+ 1.31001 6.1 1+ 1 .01 3.7 ?fa 1p<.051 Both 12.81n1a1 13.01+ 0.1601+ 1.11 17.71+1.71 2.3 ?fa 1p<.011 23 4.91171 38.3 ?161101 SC Male 21.31n1a1 22.5 1+ 1.21 22.31.11 24. 6 1+ 2. 21 26.11+1.51 28.01+1.91 +1.8 ?fa 1p<.011 Female 5 4 [n7a171-0.7101+ 1.3121+ 0. 21 701+ 0.81 8. 4 1+ 1.41 3.4 ?fa 1p<.051 Both 15.71n1a1 15.81+0. 11 17.1 1+ 1.31 19.31+2.21 1971+ 0.41 22.6 +2.91 +2.5 ?fa 1p<.011 10 +7.01 71 +44.5%1 61 SD Male 27.6 [n1a1 26. 311 .31 27. 9 1+ 1.61 30.1 1+ 2. 21 3201+ 1.91 33. 6 1+ 1.61 1.6 ?fa 1p<.011 Female 4.2 1n1a1 5.81+1.6141+ 0 6131+ 2. 01 31-1 01 11. 31 4. 01 5.8 ?fa 1p<.011 Both 14.61n1a1 1521+ 0.61 16111721+ 0.01 18. 2 1+ 1 .01 +1.4 ?fa 1p<.011 22 3.51281 24.2 ?161311 TN Male 25.1 1n1a1 2541+ 0.31 2681+ 1.31 8 01+ 1. 21 2861+ 0 61 29. 8 1+ 1.21 1.2 ?fa 1p<.011 Female 5.4 [n1a1 631+ 0.91 6 71+ 0 4151+ 0. 81 6.9 (a 0.61 7. 6 1+ 0. 71 1.9 ?fa 1p<.051 Both 12.21nfa1 12.71+ 0.61 12.31- 0.41 13211451+ 0. 91 +1.1 ?fa1p<.011 41 2.31371 18.9 ?161361 TX Male 20.4 [n7a1 2091+ 0.51 20.4 1- 0.61.0123.1 1+ 0. 91 0.9 ?fa 1p<.051 Female 4.8 [his] 5.4 1+ 0.61 5.0 1+ 0.4121+ 0. 21 0.41 6. 4 1+ 0. 81 1.6 ?fa 1p<.051 Both 17.21n1a1 19.0 1+ 1.81 18.21? 0.2521+ 1.21 2.7 ?fa 1p<.011 5 8.01 3 46.5 ?fa 1 4 1W1 UT Male 28.21n1a1 31.1 1+ 2.91 29.4 1- 1.71 32 512.1 ?fa 1p<.051 Female 6.8 [n7a1 7.41+ 0.61 7.51+ 0.2.11 1261+ 2.01 4.4 ?fa 1p<.011 Both 13.21n1a1 16.2 (+3.01 14.91-131 16.61+1.71 18.71+ 2.11 19.71+ 1.01 2.4 ?fa 1p<.011 18 6.41 91 48.6 ?Va 1 21 VT Male 23.6 [n1a1 28. 31+ 61 24.31- 4.01 2731+ 3.01 3101+ 3. 71 3251+ 1.51 1.9 ?fa 1p<.051 Female 4.3 1n1a1 21+4 0. 91 641+ 1.31 6 61+ 0.21 7. 31+ 0. 71 761+ 0.31 3.8 ?fa 1p<.011 Both 12.81n1a1 12.71-011 1291+ 0.31 1361+ 0.71 1461+ 0. 91 1501+ 0.51 +1.2 ?fa 1p<.011 37 2.21391 17.4 ?161411 17.4 Male 21.6 1n1a1 21.31- 1 2101? 0.41 2251+ 1.51 23.61 1 2.1 2391+ 0.21 0.9 ?fa 1p<.051 Female 5.3 [n1a1 5.21? 0.11 591+ 0.71 5.61? 0.31 6. 4 1+ 0 81 691+ 0.51 1.8 ?fa 1p<.051 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (nia) 15411781+ 1.0) 1.1 5511:1405) 24 2.8133) 18.8 c15137) WA Male 24.? [nfa) 2521+ 0 5) 2527.1 1+ 1.1) 13.5 '14 Female 5.0 (We) 8. 4 1+ 0 8) 81+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% 11 +58113) +37.1 i?15114) WV Male 27.2 (nia) 3011+ 2. 9) 5.81 1.+11% Female 5.3 (Na) 551+ 0.581+ 2. 2) 3.7 ?it: Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps01) Female 5.1 (nia0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We.2) 47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (ma) 521+ 0.5) 41+ 5 52) 1+ 1.4) 1251+ 1. 9) 3.2 '14. 1p<.01) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 :Decrease 1.0% I:1lnr:rease 59% - 18.3% - Increase - 29.3% - Increase 31.9% - 37.4% - increase 316% - 57.8% Rev 3.23.18, e-clearance Short Title: Vital Signs: Identifying Increasing Trends in State Suicide Rates and Exploring Suicide's Multiple Contributing Circumstances Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, F'hD;l Scott R. Kegler, Keming Yuan, M541 Kristin M. Holland, Asha Z. Ivey?Stephenson, Alex E. Crosby, M01 Background: Suicide rates in the United States have risen nearly 30% since 1999. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in ageeadjusted suicide rates among people aged 2:10 years, by state and sex, across six consecutive three?year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, DC. Data from the National Violent Death Reporting System covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Among decedents in 54.0% did not have a known MHP. Among decedents with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs 49.7%) and recentfimpencling crises (32.9% vs were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2016. A variety of circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age-adjustedh occurred in the United States among people 210 years old (1). Between 1999 and 2015, suicide rates increased across both sexes, racial/ethnic groups, and urbanization levels Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self?ha rm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The Notionoi Strotegyfor Suicide Prevention{N53P) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal). Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on clinically?oriented interventions focused on people with mental health problems (MHP) or people who have attempted suicide Other contributing circumstances include social isolation, relationship problems, economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor c0ping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this 1 Rev 3.28.18, e-clearance study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi-Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 2:10 years only, as attributions of suicidal intent in younger children are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {internationoi Ciossificotion of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census BureauiNational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada {which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicidessuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.S. military. While firearms were the most common method of suicide overall and for both groups, decedents without known MHP were more likely to die by firearm {55.3% vs. 40.6%] and less likely to die by 100I 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 Rev 3.23.18, e-clearance (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems (a0R=0.7, 95% While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP versus those with known lv?iHP had significantly greater likelihood of any relationship problem/loss (45.1% vs. specifically intimate partner problems (30.2% vs. (12.5% vs. and recently perpetrating interpersonal violence vs. They were also more likely to have experienced any life stressors (54.2% vs such as criminalalegal problems (10.7% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner and physical health problems were the most common types for both groups and did not differ between them. Decedents without known MHP had significantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be released from a correctional facility (25.2% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MH P, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%) and prior suicide attempts (10.3% vs. More than 1 in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant increases in suicide rates and 25 states experienced rate increases of more than 30%. Rates increased significantly among males, (in 34 states) and females (in 43 states). This finding is consistent with prior research showing a decreasing gender gap in male-female suicide rates between 1999- 2014 (3). Additional research into the specific causes of these trends is necessary. In the meantime, provides insight into suicide circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide ?is not caused by a single factor,? however, suicide prevention research and practice is heavily weighted towards clinically?oriented activities (6). In contrast, the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship issues and other life stressors such as criminal-legal, housing, and health problems. They also experienced more recent/impending crises which may point to suicides that took place with minimal deliberation (10). In addition to educating about how to respond to signs or disclosures of suicide risk, broader prevention to address non-clinical risks is critical and can include strengthening economic supports through housing stabilization policies, household financial support); teaching coping and problem?solving skills early in life to bolster healthy relationships over the life course; and promoting social connectedness to increase one's sense of belonging. 126 1.27r 128 129 130 131 132 133 134 135 136 .137 138 139 140 141 142 143 144 14s 146 147 148 149 150 151 152 [53 154 155 156 157 158 159 161] 161 162 163 164 165 166 167 163 Rev 3.28.13, e-clearance The above strategies are also applicable to people with MHP who, in the current study struggled more with job/financial problems, bereavement issues, and family crises, and who commonly experienced argumentsiconflicts, intimate partner, and health problems. These findings support a body of literature indicating increased vulnerability among people with MHP to socio?economic and health problems In the clinical realm, two-thirds of people with MHP had a history of any treatment and 54% were in treatment when they died. This suggests the need for broader implementation of evidence-based treatment collaborative care, cognitive-behavioral therapy], greater access to behavioral health providers (especially in underserved areas], and expanded integration of physical and behavioral health within healthcare systems, especially through care transitions In addition to the abovementioned strategies, other components of a comprehensive approach include: creating protective environments by reducing access to lethal means medications, firearms) among people at risk, creating workplace policies encouraging employee help-seeking and connectedness, and zoning to reduce alcohol-outlet density; better identifying people at risk; supporting friends/family after a suicide has taken place and assuring safe reporting by the media in order to prevent suicide contagion {121. Some states, such as Colorado, are underway planning such a comprehensive approach to suicide prevention (13). These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths overtime [potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27' states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more inndepth interviews with next?of-kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem with many associated risk factors. States and communities can use data to better understand their suicide problem and can use resources such as Preventing Suicide: a Technicai Package of Poiicies, Programs, and PracticesllZ) to implement evidence-based comprehensive suicide prevention to save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC 4 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 18.9 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 Rev 3.28.18, e-clearance References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Cu rtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2015. 4. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 2. Torguson K, O?Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnovv Mj, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow EN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1} Stone_Suicicle Vital Signs MMWR 3.27.18 (Tables 2&3} Word Count: 179971800 ONUt-lh- OJ Mom?sRev 4.2.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Background: Suicide rates in the United States have risen nearly 30% since 1999. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods {1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. Among decedents in 2? states, 54.0% did not have a known MH P. Among people with circumstance information, several circumstances, including relationship problems/loss {45.1% vs life stressors (54.2% vs and recentlimpending crises {32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides {56/100,000 [age?adjustedi) occurred in the United States (US), among people 210 years old Between 1999 and 2015, suicide rates increased across both sexes, racialfethnic groups, and urbanization levels 3). Suicide is the 10"h leading cause of death and is one ofjust three leading causes that are increasing 4). Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The National Strotegyfor Suicide Prevention{NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familyirelationship, community, and societal). Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on clinically-oriented interventions focused on mental health problems and preventing re?attempts Apart from MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor coping and problem?solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this Rev 4.2.18, e-clearance study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi-Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 2:10 years only, as attributions of suicidal intent in younger children are variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {internationoi Ciossificotion of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census BureauiNational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada {which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicidessuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US military. While firearms were the most common method of suicide overall and for both groups, decedents without known MHP were more likely to die by firearm {55.3% vs. 40.6%] and less likely to die by 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 Rev 4.2.13, e-clearance (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems (a0R=0.7, 95% While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without known MHP versus those with known lv?iHP had significantly greater likelihood of any relationship problem/loss (45.1% vs. specifically intimate partner problems {30.2% vs. (12.5% vs. and recently perpetrating interpersonal violence vs. They were also more likely to have experienced any life stressors (54.2% vs such as criminalalegal problems (10.7% vs. or evictionfloss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. Among those with crises, intimate partner problems were the most common types for both groups and did not differ between them. Similarly, physical health problems (23.2% and 21.4%) and job/financial problems (15.6% and 16.3%) did not differ between people without MHP and those with MPH, respectively, and were instead common to both groups. Decedents without known MHP had significantly lower odds of recent release from any institution 95% 0204?05), but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MH P, compared to those with MH P, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%) and prior suicide attempts (10.3% vs. More than one in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male? female suicide rates between 19992014 (3). Additional research into the specific causes of these trends is necessary. Fortunately, data from the 27 states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor; (5) however, suicide prevention research and practice is heavily weighted towards clinically-oriented activities focused on MHP (6). The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal?legal matters, evictionfloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation 124 125 126 127 128 129 130 13] 132 I33 134 135 136 137' 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 Rev 4.2.13, e-clearance Among people with MHP, two-thirds had a history of mental health and/or substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor?patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems to integrate physical and behavioral health that better support suicide prevention and patient safety, especially through care transitions (11). Study findings indicate that people with known MHP also experienced other life stressors such as job/financial, relationship, and/or physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor social, health, and economic outcomes Together, these results underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk. Other strategies include creating protective environments le.g., reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent ofinformant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next~of-kin often see greater attributions to mental disorders (14), however many methodological variations across studies exist it is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices (11) to better understand their suicide problem, prioritize evidence?based comprehensive suicide prevention, and save lives. Acknowledgments 165 166 167 168 169 170' 171 172 173 174 175 176 177 178 179 180 181 182 183 184 I85 186 187 188 189 I90 191 I92 193 194 195 196 197 I98 199 200 201 202 203 204 205 206 207 208 209 210 211 Rev 4.2.18, e-clearance The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, 770-488-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System IWISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age GroUp, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999~2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and ofthe National Action Alliance for Suicide Prevention. 2812. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 2010;14111124-34. 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Simon TR, Swann AC, Powell KE, Potter LB, Kresnow Mj, D'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide and Life-Threatening Behavior 11. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 12. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 13. Caine ED, Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 14. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Soc 2013;59161545-54. 15. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 10. Attachments: Stone_Suicide Vital Signs MMWR 4.2.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.2.18 (Tables 2&3} Word Count: 187971800 Rev 4.5.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, PhD;l Scott R. Kegler, Keming Yuan, Kristin M. Holland, PhD,?ll Asha Z. lvey-Stephenson, Alex E. Crosby, I?lel Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) arejust one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three-year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. l_n 2015,_more than half of decedents in 2? states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedi) occurred in the United States (0.5.), among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racial/ethnic groups, and urbanization levels (2, 3). Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing (1, 4). Additionally, rates of Emergency Department visits for nonfatal self-ha rm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self?harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The National Strategy for Suicide Prevention(NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal). Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts are largely clinically-oriented, focused on treating mental health problems (MHP) and preventing re-attempts (6). Apart from MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor coping and problem?solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes Rev 4.5.18, e-clearance state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi-level comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 210 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records {internationai Ciassi?cation of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from US. Census Bureaquationai Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three?year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology}, and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and raceg?ethnicity. RESULTS {665] The most recent overall suicide rates (representing 2014-2016} varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada {which had a consistently high rate throughout}, with absolute increases ranging from +0.8 {Delaware} to +8.1 {Wyoming} per 100,000. Percentage increases in rates ranged from +5.53% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states} and females {43 states}, as well as for the U.5. overall (Table Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males and females Suicide decedents without known MHP were compared to those with MHP in 2? states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5} and racial/ethnic minorities {08 range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide {adjusted odds ratio 2.9, 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the U.5. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm {55.3% vs. 40.6%} and less likely to die by Rev 4.5.13, e-clearance 84 {25.9% vs 31.3%] or poisoning [10.4% vs 19.8%) than those with known MHP. [Comment In?ll! I like the upstream 35 focus. What do you think? 85 These differences remained significant in the adjusted models. 36 Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology 8? results. decedents without known l'leP were less likely to test positive for any substance overall 95% 88 but more likely to test positive for alcohol laOFi=1.2, 95% I i 89 All suicide decedents with lleP (reason and approximately 85% without known lleP had available i 90 circumstances information {Table People without known MHP were less likely to have any substance abuse i 9 problems 95% While two-thirds of those with known MHP had a history of mental health i 92 or substance abuse treatment just over half were in current treatment. :1 93 Decedents without, knewn?MH-P?versus those with, known MHP, had significantly greater likelihood of any i 94 relationship problemfloss [45.1% vs. specifically intimate partner problems {30.2% vs. i 95 [115% vs. and recently perpetrating interpersonal violence vs. They 96 also were more likely to have experienced any life stressors [54.2% vs 491%], such as criminal-legal problems 9? {101% vs. or evictionfloss of home vs. and were more likely to have had a crisis within the i 93 preceding or upcoming tWo weeks {32.9% vs. All of these differences remained significant in the i 99 adjusted models. Among all peoplgwith crises, intimate partner problems were the most common types and did i IUD not differ between?Ehemhy group. Similarly, among people without versus with MHP, physical health problems i i i IOI {233% Ciel enablemstl?ehsne lease werecommenysxeeiepssdand did He! IUZ differ by group. [03 Decedents without known MHP had signi?cantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be ii5 released from a correctional facility [25.7% vs. or hospital vs. than those with a known [Dd MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this 10? group Were released from facilities. IDS Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation [23.0% vs. 441.8%] and prior suicide attempts vs. More than one in five people in llO both groups disclosed suicide intent (22.4% vs. Conclusions and Comments ll l2 From 1999?2015, 44 states saw significant increases in suicide rates and 25 states experienced substantial l3 increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and I I4 females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- [5 female suicide rates between 1999-2014 Additional research into the specific causes ofthese trends is He necessary. Fortunately, data from the 2? states participating in provides important insight into suicide I circumstances and can help states identify prevention priorities. [Iii Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention ?9 research and practice is heavily oriented towards dowostream treatment of MHP and prevention of reattempts. l2l]I Additional focus on non?mental health factors, further upstream, is essential?tia public health approach as the current study found that more than half of suicide decedents in did not have a known MHF. This 122 group suffered more fLonlreiationship problems and omeglife stressors such as criminal?legal matters, 123 eviction/loss of home, and recent or impending crises. 124 125 126 127r 128 129 13f}I 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 .149 150 151 152 153 154 155 156 157 158 159 160 161 162 .163 164 165 Rev 4.5.13, e-clearance Similarly. people with MHP also experienced other life stressors such as ioblfinancial, relationship, and/or physical health problems, These findings point to the need to both prevent the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes [11}jtwo?thirds of this group had a history of ?mental health and/or substance abuse treatment, with over half in ementtreatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapieggyTAdditionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed t?e integrateg physical and behavioral health that?betteesa?ppe?with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to asseeiated?withseieide Peeventlon?steategles?FnaySuch strategies sinclude: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physical/mental health problemsl Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state?level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative,- the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in?depth interviews with next-of?kin often see greater attributions to mental disorders (13), however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technicai Package of Paiicies, Programs, and to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Rev 4.5.13, e-clearance The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARSJ. Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. in. 2. ivey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Raceg?Ethnicity, Age Group, and Mechanism of Death - United States, 2001- 2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Ito J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action: a report of the US Surgeon General and ofthe National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchiapone M, et al. Suicide prevention strategies revisited: 10?year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. S. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders (DEM-SP): American Pub; 2013. 10. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2012:injuryprev-2017-042366. 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Barthoiow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 4.3.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.3.18 (Tables 2&3} Word Count: 18?41f1800 Rev 4.5.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates 2 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, 3 Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 4 Background: Suicide rates in the United States have risen nearly 30% since 1999. There is no single cause of 5 suicide. Mental health problems are ?us? one factor contributing to suicide. Examining state?level trends Ir 6 in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicideFreventioli: if 7? planning. '8 Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six 9 consecutive three?year periods (1999?2016}, were assessed using data from the National Vital Statistics System It} for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states I I in 2015, were used to examine contributing circumstances among decedents with and without known mental I2 health problems Results: From 19992016, suicide rates increased significantly in 44 states, with 25 states experiencing increases I4 of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. In I5 2015, more than half of decedents in 2? states did not have a known Among people with In circumstance information, several circumstances, including relationship problemsfloss [45.1% vs life stressors {54.2% V5 and recentfimpending crises [32.9% vs 26.0%] were significantly more likely among IS those without a known MHP than decedents with MHP, but were common across groups. l9 Conclusions: Suicide rates increased significantly across most states from 1999-2015. Various circumstances 20 contributed to Suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach 2 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help so 23 after a suicide occurs. 24 INTRODUCTION 25 BACKGROUND AND PURPOSE 26 In 2016, nearly 45,000 suicides {15.6f100,000 [age-adjustedn occurred in the United States among people 2? a10 years old Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and 28 urbanization levels Suicide is the 10"? leading cause of death and is one of just three leading causes that 29 are increasing Additionally, rates of Emergency Department visits for nonfatal self?harm, a key risk factor 30 for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 3 nation more than $69 billion in direct medical and work loss costs 32 The Notional Strotegyfor Suicide PreventioanSSP} calls for a public health approach to suicide prevention 3-3 less tarenecessitates] meets 34 an approe?ch'uhdei'scoresthaf suicidefis rarely causedqby advisirfgl-e factor, but rathef: is n'iulti-deterinined. 35 [Despite the NSSP guidance, suicide prevention efforts are largely clinically-oriented, focused on treating mental I so health problems and preventing resattempts [ELEpaH?mihemiwmf? and prior attempts, other 3? contributing circumstances to suicide include social and economic problems, access to lethal means leg, 38 substances, firearms, bridges] among people at risk, and poor coping and problem-solving skills Expanded 39 awareness of these additional circumstances contributing to suicide risk and action to address them, can help 40 reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this 1 i +ol I ,{Comment [2:119]: Maybe take out ?1ust' Comment Iza?i]: From Jane: Suicide rates in the United States haVe risen nearly 30% since 1999. There is no single cause of suicidal behavior. Mental health problems among suicide decedents are often identified through health care visits, and individuals seeking mental health services are at elevated risk for suicide. However, many other factors can contribute to suicide risk. Examining state-level trends and the multiple contributing circumstances to suicide can inform comprehensive ,state suicide prevention planning. Comment [11119]: From lane: Apart i thoughts and behaviors include inadequate coping and problem? solving skills, inadequate support i resources, social and economic Expanded awareness of these to suicide risk and ta king action to It'l- however, unable to find any state or of comprehensive, coordinated, and I'd and prior attempts, other ,or including MHP from MHP and prior attempts, other circumstances contributing to suicidal problems, and access to lethal means fe.g., substances, firearms, bridges} additional circumstances contributing and states reach the nation's goal of i 5' address them, can help communities i reducing suicide deaths 20% by 2025 community implementing the full range He also says this isn?t true. The field has 4 Comment Iza?]: Richard suggests using this cleared language instead, "Despite the NSSP guidance, The NSSP IMG was, effective suicide prevention efforts across all relevant settings and populations. Comment Instead of Apart from 1 contributing. we could say in addition to Rev 4.5.18, e-clearance study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi-Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationoi Ciossi?cotion of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census BureauiNational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS [665] The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicideesuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by Rev 4.5.13, e-clearance 85 [25.5% vs 31.3%] or poisoning [10.4% vs 19.8%] than those with known MHP. 86 These differences remained significant in the adjusted models. 3? Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results. decedents without known MHP were less likely to test positive for any substance overall 95% 88 E49 but more likely to test positive for alcohol laGFt=1.2, 95% 90 All suicide decedents with MHP and approximately 85% without known lleP had available 9 circumstances information [Table People without known MHP were less likely to have any substance abuse 92 problems 95% While two-thirds of those with known MHP had a history of mental health 93 or substance abuse treatment just over half were in current treatment. 94 Decedents without, versus those with, known MHP, had significantly greater likelihood of any relationship 95 problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. i 96 [12.5% vs. and recently perpetratlng interpersonal violence vs. They also were more likely i 9? to have experienced any life stressors [54.2% vs such as criminal~legal problems vs. or 98 evictionfloss of home vs. and were more likely to have had a crisis withinkheprecedingoL"_ 99 upcoming two weeks [32.5% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems Were the most common types and did not differ by IUI group. Similarly, among people without versus with MHP. physical health problems {23.2% and 21.4%] and lilI?. job/financial problems [15.6% and 16.3%] were commonly experienced and did not differ by group. In} Decedents without known MHP had signi?cantly lower odds of recent release from any institution MM 95% but among those who were recently released they were significantly more likely to be IOS released from a correctional facility [25.2% vs. or hospital vs. 33.0%] than those with a known lilo MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this [07 group were released from facilities. liltl Decedents without known MHP, compared to those with lleP, were significantly less likely to have a history of [09 suicidal ideation [23.0% vs. 40.3%] and prior suicide attempts [10.3% vs. More than one in five people in both groups disclosed suicide intent [22.4% vs. I Conclusions and Comments ?2 From 1999-2015, 44 states saw significant increases in suicide rates and 25 states experienced substantial I I3 increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and I I4 females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male? I IS female suicide rates between 1599?2014 Additional research into the specific causes ofthese trends is 16 necessary. Fortunately, data from the 22 states participating in provides important insight into suicide llir circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention l9 research and practice is heavily oriented towards treatment of MHP and prevention offeattempti l20 Additional focus on non?mental health factors, further upstream, is essential to a public health approach as 2 I the current study found that more than half of suicide decedents in did not have a known MHP. This [22 group suffered more from relationship problems and other life stressors such as criminal?legal matters, [23 evictionfloss of home, and recent or impending crises. I24 Similarly. people with lleP also experienced other life stressors such asjobl?flnancial, relationship. andfor I25 physical health problems. These findings point to the need to both prevent the conditions associated with 3 Comment In?ll: More likely to have i experienced or anticipated a crisis Comment Iza?ll: "much of suicide prevention practice has focused on identifying suicidal people in the community, so increased focus is needed on how to "move upstream and prevent people from becoming suicidal people in the first place. it .IL. 126 127r 128 129 130 131 132 133 134 135 136 .137 138 139 140 141 142 143 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160I 161 162 163 164 165 166 16?.i 168 Rev 4.5.13, e-clearance mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two-thirds of this group had a history of any mental health and/or substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor?patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physical/mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward), or decreased percentages of such deaths overtime (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included rapresent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Paiicies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. 4 169 170 1'11 172 173 174 175 176 177 178 179 180 181 182 183 184 I85 I86 I87 188 189 190 I91 192 I93 I94 195 I96 I97 198 199 200 201 202 203 204 205 206 207 208 Rev 4.5.18, e-clearance Corresponding author: Deborah M. Stone, 770-488-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-StEphenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca Ml. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy 5, Xu 1, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 201?. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchia pone M, et aI. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 201?. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 2017:injuryprev-2012-042366. 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow 8N, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo 0. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: Stone_Suicide Vital Signs MMWR 3.22.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.5.18 (Tables 283} Word Count: 1868/1180!) OJ woo-4Rev 4.5.13, e-clearance Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, MD1 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedii occurred in the United States among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10"h leading cause of death and is one ofjust three leading causes that are increasing (1, Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The National Strotegyfor Suicide Prevention{NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts are largely clinically-oriented, focused on treating mental health problems and preventing re?attempts (6). Apart from MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, and poor coping and problemrsolving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes Rev 4.5.18, e-clearance state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides recommendations for multi?Ievel comprehensive suicide prevention. METHODS Suicide rates were analyzed for people aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Classification of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS [665] The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicideesuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by 100I 101 102 103 104 105 106 107 108 109 111 112 113 114 115 116 117 118 119 120 121 122 123 124 Rev 4.5.13, e-clearance (26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems (a0R=0.7, 95% While two?thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without, versus those with, known MHP, had significantly greater likelihood of any relationship problem/loss {45.1% vs. specifically intimate partner problems (30.2% vs. arguments/conflicts (12.5% vs. and recently perpetrating interpersonal violence vs. They also were more likely to have experienced any life stressors (54.2% vs such as criminalalegal problems (10.7% vs. or evictionXloss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks (32.9% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems were the most common types and did not differ by group. Similarly, among people without versus with MHP, physical health problems (23.2% and 21.4%) and jobffinancial problems (15.6% and 16.8%) were commonly experienced and did not differ by group. Decedents without known MHP had significantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. or hospital (43.7% vs. 33.0%) than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MHP, compared to those with MH P, were significantly less likely to have a history of suicidal ideation {23.0% vs. 40.8%) and prior suicide attempts (10.3% vs. More than one in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male- female suicide rates between 1999?2014 (3). Additional research into the specific causes ofthese trends is necessary. Fortunately, data from the 2? states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor; (5) however, suicide prevention research and practice is heavily oriented towards treatment of MHP and prevention of reattempts. Additional focus on non-mental health factors, further upstream, is essential to a public health approach (10), as the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship problems and other life stressors such as criminal?legal matters, eviction/loss of home, and recent or impending crises. Similarly, people with MHP also experienced other life stressors such as jobjfinancial, relationship, and/?or physical health problems. These findings point to the need to both prevent the conditions associated with 3 125 126 127 128 129 131} 131 132 133 134 135 .136 137 138 139 140 141 142 143 145 146 147V 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 Rev 4.5.13, e-clearance mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two-thirds of this group had a history of any mental health and/or substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor?patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physical/mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (M D, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent [potentially biasing reported suicide rates downward), or decreased percentages of such deaths overtime (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included rapresent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist It is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. However, any lack of awareness of decedent MHP suggests the importance of addressing the broad range of contributing circumstances. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Paiicies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. 4 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 1% 197 198 199 200 201 202 203 204 205 206 207 Rev 4.5.13, e-clearance Corresponding author: Deborah M. Stone, 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atia nta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca Ml. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy 5, Xu 1, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wesserman D, van Heeringen K, Arensman E, Sarchia pone M, et 31. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 201?. 3. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed 1, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 2005;361511491-510. Attachments: Stone_Suicide Vital Signs MMWR 3.27.18 (table 1.fig 1} Stone_Suicide Vital Signs MMWR 4.5.18 (Tables 2&3} Word Count: 186871800 Rev Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, F'hD;l Scott R. Kegler, Keming Yuan, Kristin M. Holland, PhD,l Asha Z. lvey-Stephenson, Alex E. Crosby, l?lel Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) arejust one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three-year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedii occurred in the United States (0.5.), among people 210 years old Between 1999 and 2015, suicide rates increased across sexes, racial/ethnic groups, and urbanization levels Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing Additionally, rates of gEmergency dDepartment visits for nonfatal self-harm. a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The Notionoi Strotegyfor Suicide Prevention(NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familyfrelationship, community, and societal). Such a comprehensiven approach underscores that suicide is rarely caused by any single factor, but rather, is multi- determined. Despite the NSSP guidance, suicide prevention e?eetsaselargely on identifying suicidal people, providing Roofing?treatment for mental health problems (MHP) and preventing re- attempts Apaeefsesnln addition to MHP and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, among people at risk, and poor coping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, 1 Rev assesses the multiple contributing factors to suicide, and provides resemmeedaeiens?gptions for multi-Ievel comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for people aged 210 years only, as determining suicidal intent in younger children can be difficult (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationai Ciassification of Diseases Revision, underlying-cause-of death codes X60- X84, Age-specific population estimates were obtained from US. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 2? states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coronerfmedical examiner reports (including toxicology}, and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and raceg?ethnicity. RESULTS (565} The most recent overall suicide rates (representing 2014-2016} varied four-fold, from 5.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout}, with absolute increases ranging from +0.8 {Delaware} to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from +5.53% (Delaware) to (North Dakota}, with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males [34 states} and females {43 states}, as well as for the U.S. overall (Table Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 2? states. While all decedents were predominately male (Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male (83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5} and racial/ethnic minorities {08 range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI Among adult decedents, 20.1% of those without known MHP and 15.3% of those with MPH aeepie?witheet?aad with?MH?P?espeetivelyrever served, or were currently serving, in the U.S. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%} and less likely to die by itsll? ll? IIS HQII 120 [22 123 124 125 Rev (26.9% vs 31.3%] or poisoning [10.4% vs 19.8%) than those with known lv'lHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results. decedents without known l'le-iP were less likely to test positive for any substance overall 95% such as-opioids iaDR=.9Cl 35% 04131-393 but more likely to test positive for alcohol iaOR=1.2, 95% All suicide decedents with MHP and approximately 85% without known lv'lHP [N=9,357l had available circumstances information {Table People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without. versus those with, known MHP, had significantly greater likelihood of any relationship problemfloss [45.1% vs. specifically intimate partner problems {30.2% vs. {115% and recently perpetrating interpersonal violence vs. They also were more likely to have experienced any life stressors [54.2% vs such as criminal?legal problems vs. or eviction/loss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks [32.9% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems were the most common types and did not differ by group. Similarly, among people without versus with MHP, physical health problems {232% and 21.4%] and job/financial problems [15.6% and 16.8%} were commonly experienced and?dideet?diffeeby mgroups. Decedents without known MHP had signi?cantly lower odds of recent release from any institution 95% but among those who were recently released they were significantly more likely to be released from a correctional facility [25.7% vs. {43.7% vs. 33.0%} or other facility leg, alcoholisubstance treatmentl iaDR=2.5 95% than those with a known MHP. Among decedents with known MHP who were recently released from an institution 45.7% of this group were released from facilities. Decedents without known MHP, compared to those with MHP, were significantly less likely to have a history of suicidal ideation (23.0% vs. 40.8%) and prior suicide attempts {10.3% vs. More than one in five people in both groups disclosed suicide intent [22.4% vs. Conclusions and Comments From 1999?2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male? female suicide rates between 1399?2014 Additional research into the specific causes of these trends is necessary. Fortunately, data from the 2? states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention seseaseh?aed-psac?tiee?ls heavily-often oriented towards identification of suicidal people. treatment of MHP and prevention of reattempts. Additional fonts on non?mental health factors, further upstream. is essential to a public health approach as the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship problems and other life stressors such as criminal-legal matters, evictionjloss of home, and recent or impending crises. 3 Comment IHDEH: ivvould add in th r? for opioids as it is timely and relevant Bi] 126 127 128 129 130 131 132 133 134 135 136 .137 138 139 140I 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 Rev Similarly, people with MHP alsag?glexperienced relationship problems and other life stressors such as physical health problems. These findings point to the need to both h_elp people manage prevent the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two-thirds of this group had a history of any mental health and/or substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physicalfmental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, UT) might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time [potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next?of-kin often see greater attributions to mental disorders however many methodological variations across studies exist (14). it is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known MHP suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments 167 168 169 170 171 172 173 174 175 176 177 178 179 181 182 183 184 I85 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 Rev Ayala?Wee The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/'Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017'. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, van Heeringen K, Arensman E, Sarchia pone M, et a1. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. in. Washington, 2017. S. Crepeau?Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub,- 2013. 10. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 20135916154564. 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: MMWR Tables V1 4.10.18 3pm [35 Word Count: 1&853/1800 OJ woo-4Rev 4.10.18 Short Title: Vital Signs: Contributing Circumstances to Suicide and Increasing Trends in State Suicide Rates Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Background: Suicide rates in the United States have risen nearly 30% since 1999. Mental health problems (MHP) are just one factor contributing to suicide. Examining state-level trends in, and the multiple contributing circumstances to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999-2016), were assessed using data from the National Vital Statistics System for 50 states and Washington, D.C. Data from the National Violent Death Reporting System, covering 2? states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems (MHP). Results: From 1999-2015, suicide rates increased significantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased significantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known MHP. Among people with circumstance information, several circumstances, including relationship problems/loss (45.1% vs life stressors (54.2% vs and recent/im pending crises (32.9% vs 26.0%) were significantly more likely among those without a known MHP than decedents with MHP, but were common across groups. Conclusions: Suicide rates increased significantly across most states from 19992016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence~based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides [age?adjustedii occurred in the United States (US), among people 210 years old (1). Between 1999 and 2015, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10"h leading cause of death and is one ofjust three leading causes that are increasing (1, Additionally, rates of emergency department visits for nonfatal self?ha rm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The National Strotegyfor Suicide Prevention{NSSP) calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, familylrelationship, community, and societal}. Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather, is multi- determined. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal people, providing treatment for mental health problems (MHP) and preventing re?attempts (6). In addition to and prior attempts, other contributing circumstances to suicide include social and economic problems, access to lethal means substances, firea rms} among people at risk, and poor coping and problem?solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them, can help reach the nation?s goal of reducing suicide rates 20% by 2025 (7). To assist states in achieving this goal, Rev 4.10.18 this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors to suicide, and provides options for multi~leve comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for people aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (international Classi?cation of Diseases 10*? Revision, underlying-cause-of death codes X60- X84, Y87.0, Age-specific population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics and circumstances of suicide decedents 210 years old, with and without known MHP, were compared in the 27 states with complete data participating in National Violent Death Reporting System in 2015. defines MHP as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception of alcohol and other substance dependence, which are captured separately in aggregates data from three primary data sources: death certificates, coroner/medical examiner reports (including toxicology], and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racelethnicity. RESULTS The most recent overall suicide rates {representing 2014-2016] varied four-fold, from 6.9 to 29.2 (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except Nevada (which had a consistently high rate throughout], with absolute increases ranging from +0.8 {Delaware} to +8.1 [Wyoming] per 100,000. Percentage increases in rates ranged from +5.53% [Delaware] to +57.6% (North Dakota], with increases of more than 30% observed in 25 states {Table 1, Figure Modeled suicide rate trends indicated significant increases for 44 states, for males {34 states] and females {43 states], as well as for the U.S. overall (Table Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC Suicide decedents without known MHP were compared to those with MHP in 27 states. While all decedents were predominately male [Table 2; 76.8%] and non-Hispanic white those without known MHP, relative to those with MHP, were more likely male {83.6% vs. 68.8%; odds ratio 95% CI 2.2-2.5] and racial/ethnic minorities range: Suicide decedents without known MHP also had significantly greater odds of perpetrating homicideesuicide {adjusted odds ratio 2.9, 95% CI 2.23.8]. Among adult decedents, 20.1% ofthose without known MHP and 15.3% of those with MPH ever served, or were currently serving, in the US. military. While firearms were the most common method of suicide used overall and for both groups, decedents without known MHP were more likely to die by firearm (55.3% vs. 40.6%] and less likely to die by 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 Rev 4.10.18 {26.9% vs 31.3%) or poisoning (10.4% vs 19.8%) than those with known MHP. These differences remained significant in the adjusted models. Decedents without known MHP were less likely to receive toxicology testing. Among those with toxicology results, decedents without known MHP were less likely to test positive for any substance overall 95% Such as opioids 95% but more likely to test positive for alcohol 95% All suicide decedents with MHP and approximately 85% without known MHP had available circumstances information (Table 3). People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. Decedents without, versus those with, known MHP, had significantly greater likelihood of any relationship problemfloss {45.1% vs. specifically intimate partner problems (30.2% vs. arguments/conflicts (12.5% vs. and recently perpetrating interpersonal violence vs. They also were more likely to have experienced any life stressors [54.2% vs such as criminal~legal problems vs. or evictionfloss of home vs. and were more likely to have had a crisis within the preceding or upcoming two weeks {32.9% vs. All of these differences remained significant in the adjusted models. Among all people with crises, intimate partner problems were the most common types and did not differ by group. Similarly, among people without versus with MHP, physical health problems (23.2% and 21.4%) and jobffinancial problems {15.6% and 16.8%) were commonly experienced by both groups. Decedents without known MHP had significantly lower odds of recent release from any institution 95% (2120.405), but among those who were recently released they were significantly more likely to be released from a correctional facility (25.7% vs. hospital [43.7% vs. or other facility alcoholfsubstance treatment) 95% than those with a known MHP. Among decedents with known MHP who were recently released from an institution 46.7% of this group were released from facilities. Decedents without known MH P, compared to those with MH P, were significantly less likely to have a history of suicidal ideation {23.0% vs. 40.8%) and prior suicide attempts (10.3% vs. More than one in five people in both groups disclosed suicide intent (22.4% vs. Conclusions and Comments From 1999-2016, 44 states saw significant increases in suicide rates and 25 states experienced substantial increases in suicide rates of more than 30%. Rates increased significantly among males, in 34 states, and females, in 43 states. This finding is consistent with prior research showing a decreasing gender gap in male? female suicide rates between 1999-2014 Additional research into the specific causes of these trends is necessary. Fortunately, data from the 2? states participating in provides important insight into suicide circumstances and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor; however, suicide prevention is often oriented towards identification of suicidal people, treatment of MHP and prevention of reattempts. Additional focus on non?mental health factors, further upstream, is essential to a public health approach as the current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more from relationship problems and other life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. 125 126 127 128 129 130 131 132 133 134 135 .136 13?? 138 139 140 141 142 143 144 145 146 147r 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 Rev 4.10.18 Similarly, people with MHP often experienced relationship problems and other life stressors such as job/?financial and/'or physical health problems. These findings point to the need to both help people manage the conditions associated with mental health problems in the first place, and to support people with known MHP to decrease their risk of poor outcomes Two?thirds of this group had a history ofany mental health andfor substance abuse treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. Additionally, greater access to behavioral health providers in underserved areas is needed, as is expansion of healthcare systems needed that integrates physical and behavioral health with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include: strengthening economic supports housing stabilization policies, household financial support]; teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support; and identifying and better supporting people at risk Veterans, people with physicalfmental health problems) (121. Other strategies include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention These findings have at least three limitations. In the state?level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time [potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Third, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often see greater attributions to mental disorders however many methodological variations across studies exist it is likely that some people without known MHP in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known MHP suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Rev 4.10.18 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Centerfor Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsviile, 2016. 4. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 201?. 5. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wesserman D, van Heeringen K, Arensman E, Sarchia pone M, et al. Suicide prevention strategies revisited: 10-year systematic review. The Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. in. Washington, 2017. 8. Crepeau-Hobson F. The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Injury prevention 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Attachments: MMWR TablesgV1_4.10.18_4.30pm_DS Word Count: 1903} 1800 Rev 3.19.18 v3 Short title: Vital Signs: Increasing Trends in State Suicide Rates and Contributing Circumstances Deborah M. Stone, ScD,?l Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Keming Yuan, Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, M01 Structured abstract [2561250 words?this word count is not included in the 1800 max for the remainder] Background: Overall suicide rates have been rising in the United States. Examining state-level trends in suicide and its multiple contributing circumstances, can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates, by state and sex, among people aged 210 years, were assessed using data from the National Vital Statistics System. Changes in rates were examined across six consecutive three?year periods from 1999-2016. The National Violent Death Reporting System {2015), covering 27 states, was used to examine the precipitating circumstances among suicide decedents with and without known mental health problems (MHP). Results: Forty?four states saw statistically significant suicide rate increases from 1999?2016. In 25 states, rates increased by 30% or more. Male suicide rates increased significantly in 34 states while female rates increased significantly in 43 states. People with and without known MHP had both differing and similar circumstances precipitating suicide. Several circumstance, such as any relationship problems/loss [39.6 and 45.1%, ,o 5 any life stressorsfloss (49.7 and 54.2%, 5 .01), and recent crises (26.0 and 32.9%, 5 .01), respectively, were more likely among those without known MH P, but were common across groups. Conclusions: Suicide rates rose significantly across most states from 1999-2016. Varied circumstances beyond MHP alone contributed to suicides among people with and without known MH P. Implications for Public Health Practice: States can use a comprehensive public health approach based on the best available evidence to prevent suicide risks before they occur, identify and support people already at risk, prevent-reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {259 words) In 2016, nearly 45,000 suicides (15.6f100,000) occurred in the United States (US), among people 210 years Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 101th leading cause of death and is among the only leading causes to be increasing Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, increased more than 40% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The Notional Strotegyfor Suicide Prevention(NSSP) calls for a public health approach to suicide with prevention efforts spanning across multiple levels individual, family/relationship, community, and societal}, of the social ecology. Such an approach underscores that suicide is rarely caused by any single factor alone, but rather, is multi?determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems (MHP) Other associated risk factors include social and economic problems, access to lethal means substances, firearms, bridges) among people at risk, poor coping and problemssolving skills, and prior suicide attempts, among others Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state?specific Rev 3.19.18 v3 trends in suicide rates, assesses the multiple factors associated with suicide, and provides recommendations for multi-Ievel comprehensive suicide prevention. METHODS (260 words) Suicide rate estimates and trend analyses were calculated for those aged 10 years and older [because of variability in attributions of suicidal intent, younger children are excluded} Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (internationoi Ciassbficotion of Diseases 16"" Revision underlying-cause-of death codes X60-X84, Y87.0, U03). Age-specific population estimates were obtained from U.S. Census Bureau/National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods from 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Data from 2015 from the 2? states with complete data participating in the National Violent Death Reporting System were used to compare the characteristics among suicide decedents with and without known current mental health problems (MHP). MHP are defined in as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition except alcohol and other substance use disorders {captured separately]. aggregates data from three primary data sources: death certificates, coronerimedical examiner reports (including toxicology}, and law enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and racefethnicity. RESULTS (595 words) The most recent overall suicide rates (representing 2014-2016) ranged from 6.9 (District of Columbia} to 29.2 (Montana) per 100,000 persons per year, a four-fold difference (Table Across the entire study period, rates increased in all but one state (Nevada), with absolute increases ranging from +0.2 (Delaware) to +8.1 (Wyoming) per 100,000. Percentage increases in rates ranged from (Delaware) to +57.6% (North Dakota), with percentage increases of at least 30% observed in half of all states. Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U.S. overall (Table 1). By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. Nationally, the model?estimated AAPC for the overall suicide rate was By sex, the national AAPC was for males and +26% for females. Suicide decedents with and without known MHP were compared. While both groups were predominately male and non-Hispanic white, suicide decedents without known MHP were more likely to be male (83.6% vs. 68.8%; p501), and racial/ethnic minorities (odds ratio range: 1.2-2.1; 95% Cl range 1.3) They also had significantly greater odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% CI of firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI Fifteen percent of those with known MHP and 20% without ever served in the U.S. military. 100 101 102 103 104 105 106 107 108 109 110I 111 112 113 114 115 116 117 118 119 120 121 122 123 Rev 3.19.18 v3 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without MHP {19.8% vs. 10.4%; 5 .01), most frequently by over?thescounter/otherwise unclassified drugs opioids antidepressants or benzodiazepines All suicide decedents with known MHP and approximately 35% without had precipitating circumstances information. People with MHP were more likely to have any substance abuse problems (31.6% vs. 25%, 5 .01). While two-thirds of those with known MHP had a history of MH or substance abuse treatment just over half were in current mental health treatment at the time of their deaths. Decedents without known MHP, versus those with MHP had a greater likelihood of any relationship problem/?loss {45.1 and 39.6%, 1: .01), specifically intimate partner problems {30.2 and 24.1%, 5 .01), arguments/conflicts and 13.6%, 5 .01), and recently perpetrating interpersonal violence {3.0 and .01). They were also more likely to have experienced other life stressors, such as criminal legal problems (10.7 and 5 .01) or evictionfloss of home (4.3 and 5 .01), and they were more likely to have had a crisis within the preceding or upcoming two weeks (32.9 and 26.0%, Among both groups, the most common crises were intimate partner and physical health problems. Decedents without known MHP had significantly lower odds of recent release from any institution, but among those who were recently released, those without a known MHP were more likely to be released from a correctional facility vs. p501] or hospital (43.7% vs. 33.0%, p301) than those with a known MHP. Among decedents with known MHP recently released from an institution 42.3% were released from facilities. Suicide decedents without known MHP were more likely to leave a suicide note (35.1 and 31.8%, 5 .01), while decedents with known MHP, compared to those without MHP, were more likely to have a history of suicidal ideation (40.8% vs. 23.0%, .01) and attempts (29.4% vs. 10.3%, 5 .01). Conclusions and Comments {215 words) From 1999?2016, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Rates increased significantly in 34 states among males and 43 states among females. More research into the causes of these trends is necessary One important factor associated with suicide is MH P. Nearly half of suicide decedents in had a known MHP. This group was challenged by comorbid substance abuse problems and histories of suicidal ideation and attempts While two-thirds of people with MHP had a history of MH and/or substance abuse treatment and over half were in treatment at the time of their deaths, additional support could help address the needs of this vulnerable population. This includes broader implementation of affordable and evidence?based treatments, such as doctor?patient collaborative care models and cognitivevbehavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is important, as is healthcare systems change that supports suicide prevention and patient safety through care transitions While mental health problems are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. This is noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation time, particularly among people without mental health disorders and who faced impending life crises People with known MHP also experienced other life stressors such as job andfor financial problems, relationship problems, and physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the 3 124 125 126 127r 128 129 130 131 132 133 134 .135 136 137 138 139 1413 I41 142 143 144 145 146 14? 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 Rev 3.19.18 v3 first place and the need to support people with MHP to decrease their vulnerability to poor social, health, and economic outcomes These results underscore the importance of comprehensive state suicide prevention activities that go beyond a focus on MH treatment alone. Prevention strategies may include: strengthening economic supports housing stabilization policies, household financial support], teaching coping and problem-solving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and pramoting social connectedness to increase a sense of belongingness and access to informational, tangible, emotional, and social support, as needed. Other strategies indicated by these results include creating protective environments reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help?seeking and positive social norms}, supporting family and friends after a suicide has taken place, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing a comprehensive approach to suicide prevention The study findings have at least three limitations. Related to state trend analyses, four states (MD, MA, RI, UT), rankings might have been impacted by large proportions of injury deaths of undetermined intent, or by decreased percentages of such deaths overtime, which likely include some unrecognized suicides. Second, is not yet nationally representative. This study used the most current data available which includes 27 states that represent half of the U.S. population. Third. abstractors of data are limited to data included in investigative reports. For example, medical and MH information are not captured directly from medical records but from key informants le.g., family, friends] via coroner/'medical examiner reports. Therefore, informant knowledge impacts the completeness and accuracy of the information reported, and studies including in?depth interviews with family members often see greater attributions to MH and substance abuse disorders, however methodological variation across studies exists It is likely that some people without known MHP in the current study were experiencing mental health challenges at the time of death that were either not known or reported by informants, or were not captured in primary data sources. The lack of awareness of a mental health problem suggests the importance of addressing the range of contributing circumstances. Suicide is a growing public health problem. Mental illness is an important risk factor for suicide, and is one of many requiring preventive action. Data from and resources such as CDC's Preventing Suicide: a Technicai Package of Poiicies, Programs, and Practices [11] can help states and communities better understand their suicide problem and prioritize comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Conflict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, 770-438-3942 Author Affiliations: 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 Rev 3.19.18 v3 10. 11. 12. 13. 14. 15. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atlanta, GA: National Center for injury Prevention and Control. Retrieved March 15, 2018. 2016. Ivev-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Levels by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. Kegler, 5.8., D.M. Stone, and K.M. Holland, Trends in suicide by level of urbanization?United States, 1999?2015. MMWR. Morbidity and mortality weekly report, 2017. 65(10): p. 270. Kochanek, K., et al., Mortality in the United States, 2015. iv'Ci-lS Data Brief, no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 National strategy for suicide prevention: goals and objectives for action: a raport of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. Rosenman, 5.1., Preventing suicide: what will work and what will not. The Medical Journal of Australia, 1998. 169(2): p. 100-102. Torguson, K. and A. CZl?Brien, Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. 2013': Washington, D.C. Crepeau?Hobson, F., The Autopsy and Determination of Child Suicides: A Survey of Medical Examiners. Archives of Suicide Research, 2010. 14(1}: p. 24-34. American Association, Diagnostic and statistical manual of mental disorders 2013: American Pub. Li, 2., et al., Attributable risk of and socioeconomic factors for suicide from individualwlevel, population?based studies: A systematic review. Social Science 8i. Medicine, 2011. p. 608?616. Stone, D.M., et al., Preventing suicide: A technical package ofpalicies, programs, and practice. 2017. Simon, T.R., et al., Characteristics of impulsive suicide attempts and attempters. Suicide and Life? Threatening Behavior, 2002. 32(51}: p. 49?59. World Health Organization, Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO, 2012. Caine, E.D., et 31., Comprehensive, integrated approaches to suicide prevention: practical guidance. Injurv prevention, 2017: p. injuryprev42017?042366. Hawton, K., et al., The autopsy approach to studying suicide: a review of methodological issues. Affect Disord, 1993. p. 26976. Tables and Figures {attachments} Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 182971800 LinRev 3.19.18 v3 Short title: Vital Signs: increasing Trends in State Suicide Rates and ontributing Circumstances Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, PhD;l Scott R. Kegler, Keming Yuan, Comment Sometimes it is referred to as a rcontrilziuting circumstance? and sometimes its referred lvlS;1 Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, lle1 Structured abstract {2552529150 words?this word count is not included in the 1800 max for the remainder) Comment Can a couple 9f words be added here to reflect a date Since rates Background: Over all ssuicide rates have been rising in the United States 199L?Examining state? level trends have been rising. seeword count suicide and its multiple contributing circumstances can inform comprehensive git-d- prevention lannin Comment Style consideration. Since focus on suicide and states is clear, Methods: Trends in age?adjusted suicide rates. byEstatci and sex, among people aged 210 years. were assessed these words Ml?e i? sentence usmg data from the National Vital Statistics System Changes In rates were examined across so: consecutive ?arr?W?s . . omment transition rom I to three year periods from 1999 2016 :9 National Violent Death Reporting System I ?cpven?gg? reading covers states to states, was used to examine Ina-precipitating circumstances among suicide decedents With and without known methods to the first line in the results Enental health problems ?t Comment Unclear ifthis was Results: rates inereases?anrgased ai. from 1999- supposed to be a reference or can come 2015 iI"i .14 states. In 25 states, rates increased by 30% or more. Male suicide rates increased signi?cantly In 34 out states, I_444le#Femaie rates increased significantly in 43 states?eople with and without known MHP ?hrc a I ommen I U5 5? ES lOI'i an had both differing and similar circumstances precipitating suicide. Several circumstance, such as any 7i? may not be feasible, but would suggest relationship problemsfloss [39.6 and 45.1%, any life stressorsfloss [49.7 and 54.2%, 5 and recent ?t?Hn not using MHP through the document. crises {26.0 and 32.9%, 5 respectively, were more likely among those without known l'leP. but were I a common across groups. Comment Ithink this is a Significant finding should we say more Conclusions: Suicide rates rose significantly across most states from 1999?2016. Varied circumstances beyond about it the fact that it is increasing more lv'lHP alone contributed to suicides among people with and without known MHP. Nomadic-n? 3? Comment I think edits are needed Implications for Public Health Practice: States can use a comprehensive public health approach based on the here for a few reasons. I am finding the best available evidence to prevent suicide risks before they occur, identify and support people already at risk, statements that there are differences and prevent- reattempts, and help after a suicide occurs. 5 Comment Please double check. INTRODUCTION ,3 [am getting 15.9 from WISCLARS BACKGROUND AND PURPOSE words] i? i Comment Can we be more ,e specific is it the only or is it one of only in 2016, nearly 45,000 suicides occurred in the United States among people all] years old r' 3: leading causes increasing- ?among the 15 Suicide is the 10th leading cause of death and isEamong the onlyleading causes to be increasing Cummt??t Please double Additionally, rates of Emergency Department visits for nonfatal self-harm injury, a key risk factor for suicide, wonder if ??15 might. h'fwe been increased more thank??between 2001 and 2015 Together, suicides and self-harm injuries cost the nation It, for all ages and mt ?mum to am years" 3, more than $59 billion in direct medical and work loss costs ,"[C0mment IJRVSI: Suggest identifying this as a US Surgeon General report. 1, lihs Nation? mm, .5 beta, ,0 mg, prevention With efforts spanning across multiple levels e. ., individual, family/relationship, community, and It,? ecology [which i think many think of societal}, of theEocial ecolog?." Such an approach underscores that s_uici_de Is rarel_y caused by any single factor If social, environmental and economic} alone, but rather, is multi- determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems lOther asseeiateel?conttibutingrisle factors include social and economic problems, access to lethal means leg, substances, firearms, bridges] among people at risk, poor coping and problem?solving skills. and prior suicide attempts, Expanded Ftev 3.19.18 v3 40 awareness of the additional circumstances that contribute to suicide risk apart from MHP. can help reach the 4' E?atiunts 42 state-specific trends in suicide rates, assesses the multiple contributing and 43 provides recommendations for multirlevel comprehensive suicide prevention. 44 METHODS @260 words} 45 Suicide rates were analyzed for those aged 10 years and older [because of variability in attributions of suicidal as intent, younger children are excluded] Age?specific suicide counts were tabulated based on National Vital 4? Statistics System coded death certificate records [international Classification of Diseases 10'? Revision 48 underlying-cause-of death codes Age-specific population estimates were obtained from 49 US. Census Bureaquational Center for Health Statistics bridged?race population data releases. 50 5 National and statEvlevel suicide rate estimates were calculated for six consecutive three-year aggregate periods 52 from 1999-2015. Rate estimates were age?adjusted to the U.S. year 2000 standard population and expressed per 53 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three?year data 54 aggregates, employing weighted least squares regression with inverse?variance weighting. Modeled rate trends 55 are reported in terms of average annual percentage changes 50 I 5? beta from 2015 from the 2? states with complete data participating in shaggy; National iiv'iolent Death 58 Reporting System ages 10 and older, were used to compare the characteristics among suicide I 59 decedents with and without known current menial health-nodule ms 60 disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition 6 5} except alcohol and other substance use disorders [captured aggregates data from three primary data sources: death certi?cates, coronerfmedical examiner reports [including toxicology], and law 63 enforcement reports. Decedents with and without known MHP were compared using Chi-square tests; logistic 64 regression was used to estimate adjusted odds ratios with 95% confidence intervals controlling for age 65 group, sex, and racelethnicity. as RESULTS (5121591 words] 6? The most recent overall suicide rates (representing 2014-2016] ranged from 6.9 (District of Columbia] to 29.2 (58 {Montana} per 100,000 persons per year, a four-fold difference liable ill-?Acres thesetire . increased in all but one state {luevaddlt Hith??i?i?l?i109319-535. +3.1 [Wyoming] pr" 70 per 100,000. Percentage increases in rates ranged from +59% [Delaware] to 67.6% {North Dakota}, with TI percentage increases of at least 30% observed in 25 states. 1?2 7?3 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U.S. overall {Table T4 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. 75 Nationally, the model-estimated AAPC for the overall suicide rate was By sex, the national MPO was 7's +11% for males and +16% for females. T8 Suicide decedents with and without [N=11,039l known MHP were i9 were predominately male and non-Hispanic white. .sSuicide decedents without known MHP relative to inns-r: 80 ti". known I'viI-ii- were more likely to be male [83.5% vs. p5.01,l. and racialf'ethnic minorities (odds ratio 81 range: Cl range also had 33 firearm suicide 1.6, 95% CI and of positive toxicology results for alcohol 1.2, 95% CI ad served or were 34 1-1" 13}- [19.12.954.51 Hill]. 511.959.9103}; 59.9392351353qu 85 serving sewed-in the us. military. '82 significantly greater odds of perpetrating homicide-suicide {adjusted odds ratio 2.gclearer. Tracked is an idea. i LA. f. exception is described. DEM-5 does have I if two ?Table 1? in attachments state I r: still have a signi?cantly high rate of I words so you may not be able to do this .1 Commant Is It really the 'nation?s goal? - versus the goal set by the American Foundation for Suicide ,Prevention. 1 Comment Sentence has a lot of different points. it may be easier to read if it were two sentences. Suggestion: Data from the National Violent Death Reporting System were used to compare characteristics among suicide decedents with and without known current mental health programs. Complete data from 2? states participating in in 2015 were used ,for this analyses. Comment lam confUSed by how i the alcohol and substance use disorder an alcohol use disorder diagnosis and a Lsuhstance use disorder dx. 1 Comment Some labeling of the tables and how reflected in the text need some modi?cation. There are current trends: select demographics}. The two d..l a hlilh?f d. r? Comment IBEU: Per our discussion in our last VS group meeting, should we add I a note that despite NV not increases, they Lsuicidei' x? Comment Please double check. ,Table reflects +0.3 Comment Suggest revising in ,3 order to allow in the next sentence the .i comparison group to be clear and the I order the should be considered Comment Wonder if need to be i clear here this is limited to 18+. Tracked might be a simple approach to consider. have seen MMWR want to present 96?s Comment IJRVEI: I realize this adds 4? but the "ever? in the original sentence at. In. Jk E'l'll I was block to me as a i read. Not sure ifthe additional language WW If]? [08 ?m9 Illi Ill ?2 113 ?4 H5 Ilf) 117 its i 9 l20 22 [23 IE4 IRS [20 Hey 3.19.18 y3 Although firearms were used most often in both groups, decedents with known MHP died by poisoning more than those without known MHP {19.8% vs. 10.4%: 5 most frequently by over-the-counterfotherwise f: unclassified Eirugs opioids antidepressants or benzodiazepines (25.1% . All suicide decedents with known MHP lN=9,407} and approximately 35% without had precipitating circumstances 2 on People with known MHF were more likely to have any substance abuse problems [31.6% vs. 25%, 5 While two-thirds of those with known MHP had a history of MH or substance abuse treatment just over half were in current treatment at the time of their deaths. Decedents without known MHP, versus those with known MHP had a greater likelihood of any relationship problemfloss {45,199. aadvs 39.5%, 5 specifically intimate partner problems [30.2% 24.1%, 5 andys 13.6%, pg and recently perpetrating interpersonal violence [3.0225 as.dy_s_ f, They were also more likely to have experienced other life stressors, such as bosses {197?s ass Elites ?01! estates431951902919. assesses. iiases?iessj be: assesses" likely to have had a crisis within the preceding or upcoming two weeks {32.9% antly_s_. 26.0%, Among Both grows, the most common crises were intimate partner and physical health problems. i? ?1 Comment I think the comparison group 96?s are needed. For instance, drugs (35.8% vs. I think the presentation of the 93's would be easier here and later on in the results if the p's came out and the text could simply reflect ?significant differences? or some variation of indicating in the groups ,were different. JL Comment Sometimes criminal? legal is hyphenated and sometimes it?s Ln DL .2 Comment IFCU: The all reported in this sentence appear to be only for the groUp with MHPs. Based on the sentence wording, the 96 for both groups need to Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among those who were recently released, those without a known MHP were more likely to be released from a correctional facility {25.7% vs. p501) or hospital (43.7% vs. 33.0%, p501} than those with a known Mi-lP. Among decedents with known MHP recently released from an institution 42.8% were released from facilities. Suicide decedents without known MHP were more likely than tl?io?se Without known lyii ii? to leave a suicide note i35-1i'i it?d? were more likely to have a history of suicidal ideation [40.3% vs. 23.0%, 5 .01] and attempts (29.4% vs. 10.3%, p5 .Dli? i I .rcm? N?HF-ri-r? wo-H-?iDO-lwn .-.-- Conclusions and Comments [655(215 wordsi From 1999-2015, ?14 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Rates increased significantly in 34 states among males and 43 states among females. More research into the causes of these trends is necessary MHP. This group was challenged by comorbid substance abuse problems and histories of suicidal ideation and attempts While two-thirds of people with known MHP had a history of MH andfor ?a substance abuse treatment and over half Were in treatment at the time of their deaths, additional support eeald m_ay_he p address the needs of this vulnerable population. This includes broader implementation of affordable and evidence-based treatments, such as doctor?patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is important, as is healthcare systems changethat supports suicide prevention and patient safety through care transitions While mental health problems are a significant contributor to suicide, 54% of suicide decedents in this study did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal?legal matters, evictionlloss of home, and recent or impending crises. This is noteworthy in light of findings that suggest manyi suicides and attempts occur with minimal deliberation time, particularly among people without mental health disorders and who faced impending life crises 3 r{illuminant Please double check ,be reported. JL. number. Table reflects 33.3% Comment My two cents: I think we should do more to raise attention to the fact that almost a quarter of both groups disclosed suicide intent. This is a true prevention opportunity sol think this is a common risk to highlight. Knowing space constraints, I think this is more important to highlight than the presence of a suicide note which often doesn't inform prevention. I suggest adding into Lresults and into discussion. Comment lfeel typically don?t do a lot of repeat of the specific data findings in the discussion. I think the numbers could come out here. i think it would be helpful to put this into a larger context, indicating whether this is l__consistent with other research. I also .Jk Comment lwould add the Lpercentage in parenthesis [28 I29 I30 l3l 132 I33 l34 I35 36 137 I38 I39 I40 I4I I42 I44 I45 I46 14? I48 I49 I50 I5I I52 I53 I54 I55 l56 I58 159 I60 I6l I62 I63 I64 I65 166 167r I68 169 Rev 3.19.18 v3 Heople with known MHP also Experienced other life stressors such asjob andfor financialL-pseblemss; conditions associated with mental health problems in the first place and the need to support people with known MHP to decrease theirEIulnerability}o poor social, health, and economic outcomes 2? These results underscore the importance of comprehensive In! beyond a focus on maturing} health treatment alone. Prevention strategies may include: strengthening It economic supports housing stabilization policies, household ?nancial teaching coping and ?y problem-solving skills, especially early in life to manage everyday stressors and prevent future relationship problems; and promoting social connectedness to increase a sense of belongingness and access to informational, tangibie, emotional, and social sopport,eeneeded. Other strategies indicated lay these result:- include creating protective environments leg, reducing access to lethal means among people at risk, creating organizational and workplace policies to promote help-seeking and positive social norms}, supporting family and friends after a suicide?has?taieen?piaee, and assuring safe reporting by the media in order to prevent suicide contagion [ll?Some states, such as Colorado, are planning and implementing a comprehensive we?w?e approach to The steelyfindings have at least three limitations. in the state-level analysis, rankings for four states (MD, MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent, For by decreased percentages of such deaths over timt-i which likely include some unrecognized suicides. Second, is not yet nationally representative, and incl udedes 2? states that represent hair?itiBE?iI?ui of the U5. population. Third, abstractors of data are limited to data included in invEstigative reports. For exampie. and MHmor'Ital Ill'?l? information are rial. records?eat g?from key informants leg, family, friends} via coronerfmedical examiner reports and can impacts sheared completeness and accuracy, of tfieanformation reported, and Some studies including in?depth interviews with family members It is likely that some people without known l'leP in the current study were experiencing mental health challenges at the time of death that were either not known or i?eported by menial nee-oi a dcii'essing-i-he-Fa-nge ~0i-ceni-r-i bet-ing- Suicide is a growing public health problem. [Nkntal illness is an important risk factor for suicide, and is one of many requiring preventive action. Data from and resources such as Preventing Suicide: a Technicai Package of Policies, Programs, and Practices [11] can help states and communities better understand their a suicide problem and prioritize comprehensive suicide prevention. Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Con?ict of Interest No conflicts of interest were reported. Corresponding author: Deborah M. Stone, i30-488-3942 Author Affiliations: Comment suggest pulling this out into a separate paragraph. If contained in the above, it seems to undercut the main point that many suicides don't have MPH. A new paragraph might need a lead?in sentence or maybe start with the current last sentence to frame points about common ,risks. Comment Should this he 'and?, L?or? or ?andfor?? JL. Comment Suggest using "risk of" rather than "vulnerabifity to?, as vulnerability sometimes can have negative connotations aligned with ,weakness. i- Comment IFCU: I think it is fine that the TP strategies are presented in different areas of the discussion. i don?t easily see "identify and support people at-risk?. I think that can go in the proposed paragraph immediately above when talking about common risk factors. In adding that, suggest adding so me examples of approaches leg, crisis Lintervention, gatekeeper training). Comment [17ka I am not easily following this point. Can this be stated more simply such as by variations of undetermined injury deaths, which To help wfword count, offer a couple of suggestions in next few sentences to streamline. Certainly takej'leave what you tlike and check for accuracy. ., Comment Do you mean ?reported by informants? JL Comment To me, this sentence puts the spotlight pack on mental illness and makes it likely it will be the take home message of media reporting. If that isn?t the goal of this V5, suggest framing broadly. ideas: Effective approaches are Lavailobie to prevent or ameliorate many 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 Rev 3.19.18 v3 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. 10. 11. 12. 13. 14. 15. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atianto, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. Ivey-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66118): p. 1-15. Kegler, S.R., OM. Stone, and KM. Holland, Trends in suicide by ievei of urbanization?United States, 1999?2015. MMWR. Morbidity and mortality weekly report, 2017. 65(10): p. 270. Kochanek, K., et al., Mortaiity in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Genera.f and of the Nationai Action for Suicide Prevention. 2012. Rosenman, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 169(2): 100?102. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 2017: Washington, D.C. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicai Examiners. Archives of Suicide Research, 2010. 14(1): p. 2434. American Association, Diagnostic and statistical manuai ofmentai disorders 2013: American Pub. Li, 2., et al., Attributobie risk of and socio?economic factors for suicide from individuoi?ievei, popuiation-based studies: A systematic review. Social Science Medicine, 2011. 72(4): p. 608616. Stone, D.M., et al., Preventing suicide: A technicai package ofpoiicies, programs, and practice. 2017. Simon, T.R., et a1, Characteristics of impuisive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51): 49-59. World Health Organization, Risks to mentai heaith: An overview of and risk factors. Geneva: WHO, 2012. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. Injury prevention, 2017: p. injuryprev?2017?042366. Hawton, K., et al., The autopsy approach to studying suicide: or review of methodoiogicai issues. Affect Disord, 1993. 50(23): p. 269-76. Tables and Figures (attachments) Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 1826/ 1800 Flev 3.23.18 v3.1 Short title: Vital Signs: Increasing Trends in State Suicide Rates and Contributing Circumstances among people i 2 all] years 3- Deborah M. Stone, ScD;l Thomas R. Simon P'hlIl;I Katherine A. Fowler, Scott R. Kegler, Keming Yuan, ,r 4 lv'iS:1 Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, 5 Structured abstract {2501250 words?this word count is not included in the 1300 max for the remainder] is Background: Suicide rates among people?ggtial? years have risen nearly 30% since 1999. Mental health 3' 7 problems arejust one factor contributing to suicide. Examining state-level trends in suicide and other Fl contributing circumstances can inform comprehensive state suicide prevention planning. 9 Methods: Trends in age?adjusted suicide rates among people aged 3:10 years, by state and sex, across six IO consecutive three-year periods [1999-2916], were assessed usingmirom the National Vital Statistics System my 50 states and Washington, (DC). Data from the National Violent Death Reporting System, covering 27 [2 states in 2015, Wrecipitating circumstances among decedents with and without l3 known MHP. t, l4 Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases [5 of or Rates increased significantly among males and females, in 34 and 43 states, -c 1 differing circumstances and some circumstances were common to both. Among people without known l?leP, IS compared to those with 45.1% and 39.6% had any relationship problemsfloss, 54.2% and 49.7% Comment [so k6]: I. Comment Isnk?l: ,query application. I Comment [so k6]: editors will be as well. i Commont Isnk?]: I?ve never seen the MMWR of?ce strictly enforce the stated limit of 250 words for the Abstract. Here, I think we need to either say "aged all] years? or "21o years old". Another option would be to drop the phrase ?among people aged all] years? altogether right here, as this is a general lead-in statement, and just say "suicide rates?, since the specific age range is documented in the Methods section Just below. .4 Without this, it sounds like the National Vital Statistics System can be Used like a JI. Being fussy here, as I anticipate the The state ranked 15?" had an increase of The state ranked 25?? had IQ lpg?l] had any life and 32.9% and 26.0% [pg?l] had any recent/impending crises, respectively. 20 Conclusions: Suicide rates increased significantly across most states from 1999-2015. Various circumstances 2 1 Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach 23 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 24- after a suicide occurs. i 1? INTRODUCTION . AND PURPOSE {250f250 words] I 2? In 2916, nearly 45,000 suicides Ihgdcadipstedl} o?c?curr?eo?in th_e_ll_nited_5tates 28 210 years old Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and 29 urbanization levels Suicide is the 1E]th leading cause of death and is one ofjust three leading causes to be 30 increasing Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for 3 suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the 32 nation more than $69 billion in direct medical and work loss costs 33 The National Strategy for Suicide PreventioanSSP} calls for a public health approach to suicide prevention 34 with efforts spanning across multiple levels ii.e., individual, family/relationship, community, and societal]. Such 35 an approach underscores that suicide is rarely caused by any single factor alone, but rather, is multi-determined. 36 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with 3? mental health problems Other contributing factors include social and economic problems, access to 38 lethal means substances. firearms, bridges] among people at risk, poor coping and problem-solving skills, 39 and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide contributed to suicides among people with and without known MHP. ll \an increase of JL. Comment Consider adding. Flev 3.23.18 v3.1 40 risk apart from MHP, can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state?specific trends in suicide rates, assesses the multiple contributing 4 42 factors, and provides recommendations for multi-Ievel comprehensive suicide prevention. 43 METHODS {255,i'250 words} 44 Suicide rates were analyzed for people aged :10 years only, as attributions of suicidal intent in yoUnger children 45 are variable Age?specific suicide counts were tabulated based on National Vital Statistics System coded death 46 certificate records {lntemotionoi Classification of Diseases 10?? Revision, underlying-cause?of death codes X60- 4? X34, YBIG, 003}. Age-specific population estimates were obtained from US. Census BureauiNational Center for 48 Health Statistics bridged~race population data releases. 49 50 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods I 5] 1999-2016. Rate estimates were age-adjusted to the U5. year 2000 standard population and 52 expressed per 100,000 persons per year. Age~adjusted suicide rate trends Were modeled using the same three 53 year data aggregates, employing weighted least squar? regression with inverse-variance weighting. Modeled 1. 54 rate trends are reported in terms of average annual percentage changes 55 56 Characteristics {Table 2) and circumstances {Table 3) of suicide decedents 210 years, with and without known 5? MHP, were compared in the 2? states with complete data participating in National Violent Death 3 58 Reporting System in 2015. defines MHP as disorders listed in the Diagnostic and Statistical 59 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 60 defined as MHP, and were examined in both groups. aggregates data from three primary data sources: 5: 6 death certificates, coronerfmedical examiner reports {including toxicology), and law enforcement 62 reports. Decedents with and without known MHP were compared using Chi?square tests, Logistic regression 63 analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group, sex, and (14 raceiethnicity. E. I as nesuus (5341500 words] as The most recent overall suicide rates {representing 2014?2015] varied four?fold, from 5.9 to 29.2 I I 6? {Montana} per 100,000 persons per year {Table 1). Across the study period, rates increased in all states,- except 68 Nevada {with the 9? highest current suicide rate], with absolute increases ranging :59 {Wyoming} per 100,000. Percentage increases in rates ranged from +5.51% (Delaware) to {North Dakota), I if) with increases of at?leastmore than 30% observed in 25 states. 'il ?i2 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall {Table 73 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. 7'4 Nationally, the model-estimated MP0 for the overall suicide rate was By sea, the national AAPC was 75 +11% for males and +15% for females. 76 77' Suicide decedents without known MHP {N=ll,039} were compared to those with MHP While all Til decedents were predominately male {Table 2; 25.8%} and non?Hispanic white those without known 79 MHP, relative to those with MHP, were more likely male {83.6% vs. 58.8%; adjusted odds ratio 95% Cl 80 2.2?2.5] and racialfethnic minorities {odds ratio range: Cl range Suicide til decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide 2.9, 82 95% CI of firearm suicide 1.6, 95% CI and oftesting positive for alcohol 1.2, 33 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever 34 served in the US. military. Comment Izai??l: Scott we had this as 0.2 beforei'? Can you con?rm 0.8 is correct? checked against the actual 5A5 output and against Table 1 {where It is correct]. Rev 3.23.18 v3.1 H5 Although firearms were used most often, overall decedents with known MHP died by poisoning Comment lsnk?l: so significantly more than those without known MHP [33.3% vs. most frequently by over-the- Maybe we can drop this phrase without 8? counter/otherwise unclassified drugs opioids antidepressants or benzodiazepines taking BWEV mm the int??dEd and save a few Words? Comment Much less common or somewhat less common? 88 89 All suicide decedents with known MHP and approximately 85% without MHP had available 90 circumstances information (Table People without known MHP were 30% less likely to have any substance 9 abuse problems 95% While twovthirds of those with known MHP had a history of mental 92 health or substance abuse treatment just over half were in current treatment. 93 Decedents without known MHP, versus those with known MHP had a significantly greater likelihood of any 94 relationship problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. 95 arguments/conflicts [12.5% vs. and recently perpetrating interpersonal violence vs. They 96 Were also significantly more likely to have experienced other life stressors, sueh as criminal-legal problems 9? {10.7% vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 98 the preceding or upcoming two weeks [32.9% vs. Among both groups, the most common crises were i 99 intimate partner [36.2% vs. 34.9%} and physical health problems [13.8% vs. respectively. i'I li IUU Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among those who were recently released, those without a known MHP were significantly more likely to be released I02 from a correctional facility {25.2% vs. or hospital [43.2% vs. 33.0%} than those with a known MHP. Among g' Ill} decedents with known MHP recently released from an institution 42.3% were released from ., Ill-4 facilities. Decedents without known MHP, compared to those with known I?leP, were 60% less likely to have a history of lilo suicidal ideation laDR=0.4, 95% and 70% less likely to have an attempt history 95% 10? Both groups disclosed suicide intent frequently, {22.4% vs. reSpectively. lilti Conclusions and Comments (6801700 words} ?if;ll The rise in suicide rates in the ovErall US. has been observed for many years Reporting of state-specific ID common. Similarly, while geographic patterning in suicide rates have I I frequently been reported the current findings point to a disturbing pattern of increases nationwide. ?2 Understanding the contributing circumstances of suicide is obligatory for prevention practice and decision- I IS making. I I4 Research and prevention practitioners regularly state that suicide is not caused by a single factor, however, the I IS focus of suicide research and prevention practices almost solely focus on MHP. The current study found that us more than half of suicide decedents in did not have a known MHP. This group suffered more relationship ll? problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal If) deliberation l20 Among people with l'v?lHF, two?thirds had a history of mental health andfor substance abuse treatment and over l2l half Were in current treatment. This suggests that additional supports for this population are needed to keep [22 them safe. This includes broader implementation of affordable and effective treatment modalities such as 123 doctor-patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to l24 behavioral health providers, especially in underseryed areas is needed, as is expansion of healthcare systems Rev 3.23.18 v3.1 25 needed that integrate physical and behavioral health and that better support suicide prevention and patient [26 safety, especially through care transitions ll?i Study findings indicate that people with known MHP also experienced other life stressors such as jobifinancial, Ilii relationship, andr?or physical health problems. These findings point to the need to both prevent the conditions I29 associated with mental health problems in the first place and the need to support people with known MHP to Lit} decrease their risk of poor social, health, and economic outcomes Lil These results, together, underscore the importance of comprehensive statewide suicide prevention activities I32 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I33 economic supports le.g., housing stabilization policies, household financial support}; teaching coping and l34 problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in [35 life; promoting social connectedness to increase a sensa of belongingness and access to informational, tangible, I36 emotional, and social support, and identifying and better supporting people at risk. Other strategies include creating protective environments leg, reducing access to lethal means among people at risk, creating I38 organizational and workplace policies to promote help-seeking, easing transitions into and out of work for I39 people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe I40 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention I42 These findings have at least three limitations. In the state-level analysis, rankings for four states D, MA, RI, I43 might have been impacted by large proportions of injury deaths of undetermined Intent?iaerdeereasing I44 seiei-de?ratespotentially biasing reported suicide rates downward], or decreased percentages of such deaths over 145 time biasing estimated rate trends I46 nationally representative, the 2? states included in the current study reoresent 49.5% of the US. population. Third, abstractors of data are limited to information contained in investigative reports. Therefore, the I48 extent of informant knowledge can impact data completeness and accuracy. Studies including in-depth I49II interviews with next-of-kin often see greater attributions to MHP and substance abuse disorders, however many 150 methodological variations across studies exist It is likely that some people without known MHP in the ISI current study were experiencing mental health challenges that were unknown, and hence unreported by key I52 informants. However, any lack of awareness of decedent MHP suggests, even further, the importance of I53 addressing the range of contributing circumstances. I54 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are lit? available. States and communities can use data from and resources such as CDC's Preventing Suicide: 0 [So Technicai Package of Policies, Programs, and Practices [11] to better understand their suicide problem and 157I prioritize evidence-based comprehensive suicide prevention. ISR Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I60 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. Iol I62 Con?ict of Interest No conflicts of interest vvere reported. I63 I64 Corresponding author: Deborah M. Stone, TTD-488-3942 65 I66 Author Af?liations: i i Comment [calm]: More precise explanation of the potentiai influences, at the expense of just a few added words. 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 Rev 3.23.18 v3.1 1Division of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC References: 1. 10. 11. 12. 13. 14. 15. Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System Atianto, GA: Nationai Center for injury Prevention and Controi. Retrieved March 15, 2018. 2016. Ivey-Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death - United States, 2001-2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. Kegler, S.R., OM. Stone, and KM. Holland, Trends in suicide by ievei of urbanization?United States, 1999?2015. MMWR. Morbidity and mortality weekly report, 2017. 66(10): p. 270. Kochanek, K., et al., Mortaiity in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Genera.f and of the Nationai Action for Suicide Prevention. 2012. Rosenman, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 169(2): 100?102. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 2017: Washington, D.C. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicai Examiners. Archives of Suicide Research, 2010. 14(1): p. 2434. American Association, Diagnostic and statistical manuai ofmentai disorders 2013: American Pub. Li, 2., et al., Attributobie risk of and socio?economic factors for suicide from individuoi?ievei, popuiation-based studies: A systematic review. Social Science Medicine, 2011. 72(4): p. 608616. Stone, D.M., et al., Preventing suicide: A technicai package ofpoiicies, programs, and practice. 201?. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51): 49-59. World Health Organization, Risks to mentai heaith: An overview of and risk factors. Geneva: WHO, 2012. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. Injury prevention, 2017: p. injuryprev?2017?042366. Hawton, K., et al., The autopsy approach to studying suicide: or review of methodoiogicai issues. Affect Disord, 1993. 50(23): p. 269-26. Tables and Figures {attachments} Table 1 and Figure 1.doc Tables 2 and 3.pdf Word Count: 1262/1800 Rev 3.23.18 v3.1 Short tItle ?v'Ital SIgns Increasing Trends in State Suicide Rates and Fontributing Circumstances among people 2 all) years ,p Comment IBEH: Sometimes itis . referred to as a 'contributing circumstance? and sometimes its referred 3 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott Kegler, 2 Keming Yuan, i to as a ?contributing factor'_ i wonder if its 4 M531 Kristin M. Holland, Asha Z. Ivey-Stephenson, Pht'l;1 Alex Crosby, lle1 I best to be consistent so as not to confuse i the reader that the two are different. 5 Structured abstract [2501250 words?this word count is not included in the 1800 max for the remainder] I From a plain language perspective, I . i prefer contributing factors (it Also, a Background Suleide rates among people 210 years have men nearly 39% 5mm 1999. Mental health problems I. sometimes it is referredto a just 7 IMHPI are just one factor contributing to suicide. Examining state-level trends in suicide and other contrIbutIng I contributing factor' while other timeg it is circumstances can inform comprehensive state suicide prevention planning I referred to as ?contributing risk factor?. I 9 Methods Trends In age adjusted suicide rates among people 210 years, by state and sex across six consecutive l0 three- -year perIods (1999- -2016I, were assessed using the National Vital Statistics System In 50 states and II Washington D. ID (1..) Data from the National Violent Death Reporting System, Itovering.22_ states in 2015, a [2 used to examined contributing paeei?eitat-I-ng?circumstances among decedents with and without known MHP. Also, sometime it is referred to as I ?contn?buting? and sometimes as 1 ?associated'. Should we be consistent? I I i I FOUND: CONTRIBUTING PRECIPITATING I I2: Results From 1999 2016, suicide rates increased significantly in 44 states, with 25 states eapenencmg Increases more. Rates Increased significantly among males and females, in 34 and 43 states respectively Over 4'5 [5 half [54. bf decedentspeople did not have a known _AI_'no_ng decedents with circumstance CONTRIBUTING Io InformatiorI, those without known MHP. were more likely Iall 01} than those with a MHP to have relationship FACTORS IALDN -.. all. y, Comment Itgs'il]: lthink we can say contributing circumstances a few times in '9 2% and?49' and the and then lust use Comment [1919]: lsthisstillaccurate? i I 2U recentfimpendine crises I32. 9% vs 25. but these circumstances were common across groups adiFHf?I?I'Comment [tgs9 : Some editing is needed because this read like the data examined circumstances. 7'9 Conclusmns rates Increased significantly across most states from 1999-2016 lifarious CIrcumstances 23 contributed to suicides among people with and without known MHP Comment IzaI?Jl: Switched the order here JL 24 Implications for Public Health Practice: States can use a comprehensive evidence based public health approach 75 to prevent rIsk before it occurs, identify and support people at risk prevent reattempts, and help 26 friends/family after a suicide Comment ltgs9]: I saw that Cory made a 2? INTRODUCTION good suggestion to say among those with circumstance information. If we make one point about the p?value we can save 28 anckonouno AND PURPOSE {250/st words} words The edits suggested allew US that there are differences but these are 29 In 2016 nearly 45,000 suicides occurred in the United States I, among people >10 years old 30 . Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and urbanization levels I 3 31. Suicide is the 10Lh leading cause of death and is one of just three leading causes marquees?be increasing 32 Additionally, rates of Emergency Department visits for nonfatal self harm a key rIsl< factor for sUIcIde, 33 Increased nearly 45% between 2091 and 2015 Together, suicides and self-harm injuries cost the when more 34 than $69 billion in direct medical and work loss costs gcommon for both. 35 The Notional Strategyfor Suicide PreventionI'NSSP} calls for a public health approach to suicide prevention 36 with efforts spanning across multiple levels Ii.e., individual, familyfrelationship, community, and societal]. Such I 3? an approach underscores that suicide is rarely caused by any single factor?alone, but rather, is multi-determined 38 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals With I 39 mental health problems IMHPI Other contributing faetersrcircumstances include SOCIEII and economic 1 Flev 3.23.13 v3.1 40 problems, access to lethal means substances, firearms, bridges} among people at risk, poor coping and 4 problem?solving skills, and prior suicide attempts Expanded awareness ofthe additional circumstances that 42 contribute to suicide rislt apart from can help reach the nation?s goal of reducing suicide rates 20% by i 43 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates. assesses 44 the multiple contributing factors, and provides recommendations for multi?level comprehensive suicide 45 prevention. 46 METHODS Izssizso words} 47 Suicide rates were analyzed for people 210 years only, as attributions of suicidal intent in younger children are 48 variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records [International Ciossm'cotion of Diseases 10'? Revision, underlying?cause?of death codes ,i i i 50 X34, Age-specific population estimates were obtained from U.S. Census Bureaufi?iational Center for SI Health Statistics bridged-race population data releases. 52 53 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 54 from 1999-2016. Rate estimates were age-adjusted to the U.5. year 2000 standard population and expressed per Age?adjusted suicide rate trends were modeled using the same three-year data I i 55 100,000 persons per year. I i 56 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 5 5? are reported in terms of average annual percentage changes I 60 MHF, were compared in the 2? states with complete data participating in 0005 National Violent Death 6 Reporting System in 2015. defines MHP as disorders listed In the Diagnostic and Statistical 62 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 63 defined as MHP, and were examined in both groupskit-?DRS aggregates data from three primary data sources: 64 death certificates, coronerfmedicai examiner reports I 05 reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression 66 analyses estimated adjusted odds ratios with 95% confidence intervals controlling for age group. sex, and 58 59 Characteristics [Table 2] and circumstances {Table 3} of suicide decedents 210 years, with and without known ,race/ethnicity. as RESULTS (sac/sou wordsl 69 The most recent overall suicide rates (representing 2014?2015] varied four?fold, from 5.9 to 29.2 20 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except ?i?l Nevada {with the highest current suicide rate), with absolute increases ranging from 0.8 22 {Wyoming} per 100,000. Percentage increases in rates ranged from +53% (Delaware) to 62.6% {North Dal-totalIwith increases of at least 30% observed in 25 states. 7?4 i5 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the US. overall {Table 1] 76 By sex, modeled rate trends indicated signi?cant increases in 34 states for males and in 413 states for females. Nationally, the modelvestimated MPC for the overall suicide rate was By sex, the national MPC was 7?8 +1.1?is for males and +16% for females. 79 80 Suicide deCedents without knowo MHP {hi?111,039} were compared to those with I'leP While all til decedents were predominately male (Table 2: 75.8%} and non?Hispanic white those without known 82 MHP, relative to those with MHP, were more likely male [33.6% vs. 53.8%; adjusted odds ratio laORi=2.3, 95% Cl '83 2.2?2.5] and racial/ethnic minorities (odds ratio range: 1.2-2.1; 95% Cl range - Suicide B4 decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide laDFi 2.9, HS 95% CI of firearm suicide 1.6, 95% CI and of testing positive for alcohol 1.2, 2 Comment Item]: The comment made was that it wasn't clear how we defined MHP since substance use disorders are a diagnosis in DSM-V. Changed this to be more strai htforward. Comment [2319]: Scott we had this as 0.2 before?i' Can you confirm 0.3 is correct? 9o 98 99 100 IOI IUE [05 ?36 Illlid ll?} [20 l2l l22 123 IM- l25 [as Rev 3.23.18 v3.1 95% CI Among adult decedents, and 15.3% of people without and with MHP, respectively, ever x? served in the U.S. iIItary} .4 Although firearms were used most often, overall decedents with known MHP were nlqgejj?ygdied by suffocation (31.3 vs. 26.9%l and poisoning i19.8% vs. 111.4%] signifieantiy?mere than those without known MHP fr . . . a All suicide decedents with known MHP and approximately 35% without MHP had available circumstances information {Table 3). People without known MHP were 39% ess likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. @ecedents without known MHP, versus those with known lv'lHP had a significantly greater likelihood of any relationship problemi'loss (45.1% vs. specifically intimate partner problems [30.2% vs. {115% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other iife stressors, such as criminai?legal problems (10.3% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preteding or upcoming two Weeks [323% vs. Among both groups, the most common crises were intimate partner [36.2% vs. 34.9%} and physical health problems [13.3% vs. respectively] Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among those who were recently released 15.lltral, those without a known MHP were significantly more likely to be released from a correctional facility MHP. Among decedents with known lv'lHP recently released from an institution 42.3% were released from facilities. @ecedents without known MHP, compared to those with knewn?MHF', were Ease-less likely to have a history of suicidal ideation 95% and laid?less?ii-kely?te?have?an? prior suicide attempt histew 95% Both groups disclosed suicide intent frequently, [22.4% vs. respectively] x" J, Conclusions and Comments [680/700 words) . The rise in suicide rates in the overall U.S. has been observed for many years [Reporting of state?specific trends were? and is mush '655 [Comment I think that is fine. Comment lately]: Malia wanted to know if we could say "had served or were serving? in the US. Military. Comment I think the comparison group are needed. For instance, drugs [35.3% vs. I think the presentation of the would be easier here and later on in the results if the p's came out and the text could simply reflect ?significant differences? or some variation of indicating in the groups were different. CHANGE BIC IT OPENS UP THE ISSUE OF PEOPLE WED MHP TAKING ETC Do we eyen need to give this distribution? We don?t talk about the types in the discussion .a Comment [2:119]: I kept this paragraph con?ned to percentages but did add some gadditional aC-R's in other places. 1 Comment tgs9l: We should be consistent and add this if we have the below. J5. Comment lentil]: We were asked to include disclosure of suicide intent. Also, Lchanged the formatting here a bit. Comment [tg59]: Why did you add this? it sounds like others have done it already. I think the last version of the start of the frequently been reported the current findings point to a disturbing pattern of increases nationwide. Understanding the contributing circumstances of suicide is obligatory for prevention practice and decision? making. and preventienpractitioners regularly state that suicide is not caused by a single factor, however, the focus ofsuicide research and prevention practices almost solely focus on MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionftoss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MHP, two?thirds had a history of mental health andlor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor?patiEnt collaborative care models and cognitive?behavioral therapy. Additionally, greater access to 3 Lconcltoion Was more compelling. Rev 3.23.18 v3.1 me nt lzal'i?: Is this confirmedLTom behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems If: needed that integrate physical and behaworal health and that better support surcide prevention and patient {Comment This is from the 2015 safety, especially through care transmons i: NUDRS 55 I30 Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, ?[Comment Infill: Need to consider this E. part further, I32 associated with mental health problems in the first place and the need to support people with known to Comment [29191: Cory had a great idea here so took out, ental illness is an I33 decrease their risk of poor social, health, and economic outcomes important risk factor for suicide, and is I34 These results, together, underscore the importance of comprehensive statewide suicide prevention activities I35 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I36 economic supports ie.g., housing stabilization policies, household ?nancial support}; teaching coping and 137 problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in I38 life: promoting social connectedness to increase a sense of belongingness and access to informational, tangible, 39 emotional, and social support, and identifying and better supporting people at risk. Other strategies include I40 creating protective environments le.g., reducing access to lethal means among people at risk, creating l4l organizational and workplace policies to promote help-seeking, easing transitions into and out of work for I42 people with IleP and other life challenges], supporting family and friends after a suicide, and assuring safe I43 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning I44 and implementing such a comprehensive approach to suicide prevention She said the media would probably just take away Mental illness Is an important relationship, andfor physical health problems. These findings point to the need to both prevent the conditions until}: Is this still ok? risk factor, period. I45 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, I46 might have been impacted by large proportions of injury deaths of undetermined intent decreasing l4? suicide rates), or decreased percentages of such deaths over time he. increasing suicide rates]. Second, I48 is not yet nationally representative, the 2? states included in the current study representi?i??i?s of the US. .1 I49 lpopulation, Third, abstractors of data are limited to information contained in investigative reports. I50 ?file-refore, the extent of informant knowledge can impact data completeness and accuracy. Studies including in- ISI depth interviews with next-of-kin often see greater attributions to MHPland substance abuse disorderi I52 however many methodological variations across studies exist it is likely that some people without known I I53 MHF in the current study Were experiencing mental health challenges that were unknown, and hence I54 unreported by key informa any lack of awareness of decedent MHP suggests, even further, the I55 importance of addressing the range of contributing circumstances. I56 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are I57 available States and communities can use data from and resources such as CDC's Preventing Suicide: or l5ii Technicoi ?o-dkdEe-of?olicies, Programs, and Practices [11] to better understand their suicide problem and 159 prioritize evidence-based comprehensive suicide prevention. I60 Acknowledgments l6l The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I62 Statistics, National Center for Health Statistics. CDC, for their statistical consultation. [63 I64 Con?ict of Interest No conflicts of interest were reported. [65 lot) Corresponding author: Deborah M. Stone, dsjoneEchcgov Fi?-4883942 I o? 68 Author Af?liations: Rev 3.23.18 v3.1 I69 1Division of Violence Prevention, National Center for Injury Prevention and Control, aDivision of Analysis, 170 Research, and Practice Integration, National Center for injury Prevention and Control, CDC References1140 192 193 1:has!statistics-testesenasset Centers for Disease Control and Prevention, Web-based injuryr Statistics Query and Reporting System Ationto, GA: Nationai Center for injuryr Prevention and Controi. Retrieved March 15, 2018. 2016. Nev-Stephenson, A.2., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001?2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. Kegier, S.R., 0M. Stone, and KM. Holland, Trends in suicide by ievei of urbanization?United States, 1999-2015. MMWR. Morbidity and mortality weekly report, 201?. 66(10): 220. Kochanek, It, et ai., Mortaiity in the United States, 201? 6. NCHS Data Brief; no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goats and objectives for action: a report of the US Surgeon Generai and of the National Action for Suicide Prevention. 2012. Rosen ma n, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 201?: Washington, D.C. Crepeau?Hobsan, P, The Autopsy and Determination of Child Suicides: A Survey of Medicai Examiners. Archives of Suicide Resea rch, 2010. 14(1): p. 24?34. American Association, Diagnostic and statisticai manuai of mentai disorders 2013: American Li, 2., et al., Attributabie risk of and socia-economicfactors for suicide from individuai-ievei, popuiation-based studies: A systematic review. Social Science 81 Medicine, 2011. 22(4): p. 508-616. Stone, 0.90., et al., Preventing suicide: A technicai package of paiicies, programs, and practice. 2017. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51): p. 49-59. World Health Organization, Risks to mentoi heaith: An overview of and risk factors. Geneva: WHO, 2012. Caine, 12.0., at al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. Injury prevention, 201?: p. injuryprev?ZOlT-OAHEB. Hawtan, K., et al., The autopsy approach to studying suicide: a review of methodoiogicai issues. Affect Disord, 1998. p. 259-76. 205 Table 1 and Figure 1.doc 206 Tables 2 and 3.pdf 20? Word Count: 126210300 Comment lzu?ll: Need to replace ref 10 and fix 11-15 Rev 3.23.18 v3.1 Short title: Vital Signs: Increasing Trends in State Suicide Rates and Elontributing Circumstances among people 2 210 years 3 Deborah hiirStone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Kei?ning Yuan, 4 Kristin M. Holland, Asha Z. Ivey-Stephenson, PhD;1A ex E. Crosby, M01 5 Structured abstract [2501250 words?this word count is not included in the 1300 max for the remainder] a Background: Suicide rates among people 210 years have risen nearly 30% since 1999. Mental health problems 7 are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing Fl circumstances can inform comprehensive state suicide prevention planning. '1 i 9 Methods: Trends in age?adjusted suicide rates among people 210 years, by state and sex. across six consecutive 3 IO three-year periods (1999-2016), were assessed using the National Vital Statistics System in 50 states and I [2 used to examine contributing circumstances among decedents with and without known MHP. l3 Results: From 1999-2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases I4 of 30% or more. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over [5 half did not have a known those lo without known MHP were more likely {all p501} than those with a MHP to have relationship problemsiloss [7 {45.1% vs life stressorsfloss {54.2% vs and recentfimpending crises (32.9% vs but these IS circumstances were common across groups. 19 Conclusions: Suicide rates increased significantly across most states from 1999-2015. 1liarious circumstances 20 contributed to suicides among people with and without known 2 Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach 22 to prevent suicide risk before it occurs, identify and support people at risk, prevent and help 23 friendsifamily after a suicide occurs. 24 INTRODUCTION 25 BACKGROUND AND PURPOSE {250f250 words} 26 in 2016, nearly 45,000 suicides l15.6,i100,000j occurred in the United States among people 210 years old 2? BetweEn 1999 and 2016, suicide rates increased acrc-Ss sexes, raciali?ethnic groups, and urbanization levels I 28 31. Suicide is the 10Lh leading cause of death and is one of just three leading causes that are increasing Comment Sometimes itls Lthe text and then just use circumstances. [Comment [1:119]: Is this stillaccurate? 3 Lcircumstances. -vH" Comment jmt?Jj: Switched the order here Comment tgs9j: I saw that Cory made a good suggestion to say among those with circumstance information. If we make one point about the prvalue we can save words. The edits I suggested allow us that a I referred to as a 'contributing circumstance? and sometimes its referred to as a ?contributing factor?I . i wonder if its best to be consistent so as not to confuse the reader that the two are different. From a plain language perspective, i prefer contributing factors Also, sometimes it is referred to a just ?contributing factor' while other times it is referred to as ?contributing risk factorJ. Also, sometime it is referred to as *contn?buting? and sometimes as *associated'. Should we be consistent? I FOUND: CONTRIBUTING PRECIPITATING CIRCUMSTANCES CONTRIBUTING FACTORS Ei=3 Comment Itgs9]: lthink We can say contributing circumstances a few times in Comment [tgs9j: Some editing is needed because this read like the data examined 29 Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased 30 nearly 415% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 3 billion in direct medical and work loss costs 32 The Notional Strategyfor Suicide Preventionj'NSSP} calls for a public health approach to suicide prevention 33 with efforts spanning across multiple levels li.e., individual, family/relationship, community, and societal}. Such 34 an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. 35 Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with so mental health problems Other contributing circumstances include social and economic problems, 3? access to lethal means leg, substances, firearms, bridges) among people at risk, poor coping and problem- 33 solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that 39 contribute to suicide risk apart from MHP, can help reach the nation?s goal of reducing suicide rates 20% by there are differences but these are tcommon for both. Rev 3.23.18 v3.1 40 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses 4 the multiple contributing factors, and provides recommendations for multi-Ievel comprehensive suicide 42 prevention. 43 (2551st words) 44 Suicide rates were analyzed for people 210 years only, as attributions of suicidal intent in younger children are 45 variable Age?specific suicide counts were tabulated based on National 1v?ital Statistics System coded death 4s certificate records (international Ciassrficotion of Diseases 10?? Revision, underlying-cause?of death codes 4? X34, TRIO, U03). Age-speci?c population estimates were obtained from U.S. Census Bureaufii-lational Center for 48 Health Statistics bridged?race population data releases. 50 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 5] from 1999-2016. Rate estimates were age-adjusted to the U5. year 2000 standard population and expressed per 52 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data 53 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 54 are reported in terms of average annual percentage changes (AAPCs). 55 56 Characteristics {Table 2) and circumstances (Table 3) of suicide decedents 210 years, with and without known 5? MHP, were compared in the 2? states with complete data participating in National Violent Death 58 Reporting System in 2015. defines MHP as disorders listed in the Diagnostic and Statistical 59 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 6 death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement 62 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression as analyses estimated adjusted odds. ratios with 95% confidence intervals (CI), controlling for age group, sex, and 64 race! ethnicity. 65 RESULTS (5841600 words} as The most recent overall suicide rates (representing 2014?2016) varied four?fold, from 5.9 to 29.2 6? (Montana) per 100,000 persons per year (Table Across the study period, rates increased in all states, except 68 Nevada (with the highest current suicide rate), with absolute increases ranging fromEI-O? 69 (Wyoming) per 100,000. Percentage increases in rates ranged from +53% (Delaware) to 616% (North Dakota) it) with increases of at least 30% observed in 25 states. Tl i2 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall (Table 1). 7?3 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. T4 Nationally, the model-estimated AAPC for the overall suicide rate was By sex, the national MPC was 75 +11% for males and +15% for females. 76 Suicide decedents without known MHP were compared to those with MHP While all TS decedents were predominately male (Table 2; 75.8%) and non-Hispanic white those without known it? MHP, relative to those with MHP, were more likely male (33.6% vs. 68.8%; adjusted odds ratio 95% CI 80 2.2?2.5) and racial/ethnic minorities (odds ratio range: CI range Suicide 8 decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide 2.3, 82 35% CI of firearm suicide 1.6, 95% CI and oftesting positive for alcohol 1.2, 33 35% CI Among adult decedents, 10.1% and 15.3% of people without and with MHP, respectively, ever 34 served in the USEnilitan} 1' Comment Imam]: The comment made was that it wasn't clear how we defined MHP since substance use disorders are a diagnosis in Changed this to be more strai htl?orward. Comment [2319]: Scott we had this as 0.2 before?i' IDan you confirm 0.3 is correct? Cormm-nt ham: Malia wanted to know if we could say "had served or were serving? in the U.S. Military. Comment Itgs9]: I think that is ?ne=I=Irnz null-I! IOU IUI l02 03 04 l05 106 [(13 Ifil?IZU Ill [22 l23 24 Flev 3.23.18 v3.1 Although firearms were used most often, overall {48.5%l, decedents with known MHP were more likely to die by suffocation {31.3 vs. 26.9%] and poisoning (19.8% vs. 10.4%] than those without known MHP All suicide decedents with known MHP and approximately 35% without MHP circumstances information {Table People without known MHP were less likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. [Decedents without known MHP, versus those with known MHP had a significantly greater likelihood of any relationship problemfloss [45.1% vs. specifically intimate partner problems {30.2% vs. arguments/conflicts {12.5% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other life stressors, such as criminal-legal problems (10.2% vs. or evictiony?loss of home vs. and they were more likely to have had a crisis within the preceding or upcoming two weeks [32.9% vs. Among both groups, the most common crises were intimate partner [36.2% vs. 34.9%] and physical health problems [13.3% vs. respectively] Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among those who were dealt released released from a correctional facility [25.7% vs. or hospital (43.7% vs. 33.0%] than those with a known MHP. Among decedents with known MHP recently released from an institution 42.3% were released from facilities. [Decedents without known MHP, compared to those with MHP, were less likely to have a history of suicidal ideation 95% 0:04-05] and prior suicide attempt 95% Both groups disclosed suicide intent frequently, {22.4% vs. respectively] Conclusions and Comments words] The rise in suicide rates in the overall U.5. has been observed for many years of state~speci?c trends overall and hr 59* is much less frequently been reported the current findings point to a disturbing pattern of increases nationwide. Understanding the contributing circumstances of Suicide is obligatory for prevention practice and decision? making. Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of suicide research and prevention practices almost solely focus on MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, eviction/loss of home, and recent or impending crises. This is particularly noteworthy in light offindings that suggest many suicides and attempts occur with minimal deliberation 112]. Among people with l'leP, two-thirds had a history of mental health and/or substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keEp them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and cognitive~behavioral therapy. Additionally, greater access to behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient safety, especially through care transitions Comment Ithinkthe comparison group are needed. For instance, drugs (35.8% vs. I think the presentation of the 93's would be easier here and later on in the results if the p?s came out and the text could simply reflect ?significant differences" or some variation of indicating in the groups were different. CHANGE THIS arc IT oPEns UP THE ISSUE OF PEOPLE wro MHP TAKING ANTIDEP Do we even need to give this distribution? We don?t talk about the types in the ?scussion ?1 .4 Comment Izal?J]: I kept this paragraph con?ned to percentages but did add some Ladditional aOR?s in other places. Comment tg59]: We should be consistent and add this if we have the . Comment [2am]: We were asked to include disclosure of suicide intent. Also, Lchanged the formatting here a bit. Comment Itgs9]: Why did you add this? It sounds like others have done it already. I think the last version of the Start of the Lconclusion was more compelling. 30.2% below. JL Flev 3.23.18 v3.1 I25 Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, I26 relationship, andfor physical health problems. These findings point to the need to both prevent the conditions associated with mental health problems in the first place and the need to support people with known l'viHP to IZS decrease their risk of poor social, health, and economic outcomes [29 These results, together, underscore the importance of comprehensive statewide suicide prevention activities 13G that address multiple factors associated with suicide Prevention strategies may include: strengthening Lil economic supports housing stabilization policies, household ?nancial support}; teaching coping and 132 problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in I33 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, 34 emotional, and social support, and identifying and better supporting people at risk. Other strategies include I35 creating protective environments reducing access to lethal means among people at risk, creating 136 organizational and workplace policies to promote help-seeking, easing transitions into and out of work for people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe [38 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning I39 and implementing such a comprehensive approach to suicide prevention 4li These findings have at least three limitations. In the state-level analysis, rankings for four states MA, RI, might have been impacted by large proportions of injury deaths of undetermined intent decreasing I42 suicide rates], or decreased percentages of such deaths overtime li.e. increasing suicide rates]. Second, I43 is not yet nationally representative, the 2? states included in the current study representl?i??i?o of the US I44 lpopulation. Third, abstractors of data are limited to information contained in investigativeleports. I45 Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including in- 146 depth interviews with next-of?kin often SEE greater attributions to substance abUSe disorderi however many methodological variations across studies exist [t is likely that some people without known I43 MHP in the current study were experiencing mental health challenges that were unknown, and hence I49 unreported by key informants] However, any lack of awareness of suggests, even further, the I 5U importance of addressing the range of contributing circumstances. I5I Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are I52 available States and communities can use data from and resources such as CDC's Preventing Suicide: 0 I53 Technical Package of Policies, Programs, and Practices [11] to better understand their suicide problem and l54 prioritize evidence-based comprehensive suicide prevention. 55 Acknowledgments 156 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital Statistics National Center for Health Statistics, CDC, for their statistical consultation. [58 I59 Con?ict of Interest No conflicts of interest were reported. lot) Iol Corresponding author: Deborah M. Stone, FPO-4883942 I62 163 Author Affiliations: [(54 1Division of?v?iolence Prevention, National Center for Injury Prevention and Control, ?Division of Analysis, I65 Research, and Practice Integration, National Center for Injury Prevention and Control, CDC loo References: I[Com me at Jga?ij: Is this i) {{Comment [tgs9]: This is from the 2015 NUDRS 55 II Comment [1:119]: Need to consider this I Comment [2919]: Cory had a great idea here so took out, ental illness is an ,l important risk factor for suicide, and is I one of many requiring preventive action. She said the media would probably just take away Mental illness is an important [Comment until]: Is this still ok? risk factor, period. Rev 3.23.18 1.31 169 I68 1139 ITO 2. 111 l'iz ITS 3. 194 its 4. 116 1'1? 5. ITS 179 I80 5. 181 132 15:3 134 a. 135 13-5 s. [81? 133 1a. 139 190 11. 191 12. 192 193 13. 194 [95 14. [96 15. I98 '99 Centers for Disease Control and Prevention, Web-based injury Statistics Query and Reporting System (WISQARSJ. Arianta, GA: National L?enterfor injury Prevention and Control. Retrieved March 15, 2018. 2016. Nev?Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, Race/Ethnicity, Age Group, and Mechanism of Death United States, 2001?2015. MMWR Surveill 51.1mm, Kegler. S.R., 0M. Stone, and KM. Holland, Trends in suicide by ievel of urbanization?United States, 1999?2015. MMWR. Morbidity and mortality weekly report, 2017. 66(10): 270. Kochanek, K., et al., Mortaiity in the United States, 2016. NCHS Data Brie? no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Generai and of the National Action for Suicide Prevention. 2012. Rosenma n, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. p. 100-102. Torguson, It. and A. O'Brien, Leading Suicide Prevention E??orts Unite to Address Rising Nationai Suicide Rate. 201?: Washington, D.C. Crepeau?Hohson, 13., The Autopsy and Determination of Chiid Suicides: A Survey of Medicai Examiners. Archives of Suicide Research, 2010. 14(1): p. 2434. American Association, Diagnostic and statistical manuai of mentai disorders 2013 American Pub. Li, 2., et al., Attributabie risk of and soda-economic factors for suicide from individual-ievei, population-based studies:A systematic review. Social Science Si Medicine, 2011. RM): 608-615. Stone, D.M., et al., Preventing suicide: A technical package afpoiicies, programs, and practice. 2017. Simon, T.R., et al., Characteristics of impuisive suicide attempts and otternpters. Suicide and Life- Threatening Behavior, 2002. 32(51): p. 49-59. World Health Organization, Risks to mentai health: An overview of and risk factors. Geneva: WHO, 2012. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. Injury prevention, 201?: p. injuryprev-2011-042356. Hawton, IL, et al., The autopsy approach to studying suicide: a review ofmethodoiogicai issues. Affect Disord, 1998. a. 259?76. 200 Table 1 and Figure 1:10: 20] Tables 2 and 3.pdi 202 Word Count: 1162,1?1800 Comment lzu?ll: Need to replace ref 10 and fix 11?15 3? Flev 3.23.18 V3.1 Short title: Vital Signs: lncreasing Trends in State Suicide Rates and Fontributing Circumstances of Suicide among people 210 yea rs Deborah M. Fl. Simoanth lie?thefirfeiit?fovv?le-f, Scott R. Kegler, Reming?ruan, lVlS:1 Kristin M. Holland, Asha Z. Ivey-Stephenson, Alex E. Crosby, lVlD1 Structured abstract [2501'250 words?this word count is not included in the 1800 max for the remainder] Background: Suicide rates among people 210 years have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people 210 years, by state and sex, across six consecutive three-year periods (1999-2016), were assessed using the National Vital Statistics System in 50 states and Washington, D.C Data from the National Violent Death Reporting examined precipitating circumstances among decedents with and without known MHP. Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of 30% or more. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over half inf people did not have a known 46.0% dii Groups experienced similar and differing circumstances and some circumstances were common to without known MHP, compared to those with 45.1% and 39.5% ipc.01)i had any relationship problemsfloss, 54.2% and 49.2% (p501) had any life stressorslloss and 32.9% and 26.0% (p501) had any recentlimpending crises, respectively. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {250f250 words} In 2016, nearly 45,000 suicides l15.6f100,000i occurred in the United States among people ?ew years old Between 1999 and 2016, suicide rates increased across sexes, racialfethnic groups, and urbanization levels Suicide is the 10th leading cause of death and is one of just three leading causes to be increasing Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 5,59%. between 2001 and 2015 Together, suicides and self?harm injuries cost the nation more than $59 billion in direct medical and work loss costs The Notional Strategy for Suicide PreventioanSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societall. Such an approach underscores that suicide is rarely caused by any single factor alone, but rather, is mold-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing factors include social and economic problems, access to lethal means substances, firearms, bridges] among people at risk, poor coping and problem-solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from and preventive action can help reach the nation's goal of reducing suicide rates 20Comment Isthisstillaccurate? Comment In?ll: Switched the order Comment Sometimes it is referred to as a 'contributing circumstance? and sometimes its referred to as a ?contributing factor?I . i wonder if its best to be consistent so as not to confuse the reader that the two are different. From a plain language perspective, I prefer contributing factors Also, sometimes it is referred to a just ?contributing factor' while other times it is referred to as ?contributing risk factor?. Also, sometime it is referred to as *contributing? and sometimes as *assoclated'. Should we be consistent? I FOUND: PRECIPITATING CIRCUMSTANCES (ALONEFE CONTRIBUTING FACTORS- RISK 1 Lhere Rev 3.23.18 v3.1 40 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses 4 the multiple contributing factors, and provides recommendations for multi-Ievel comprehensive suicide 42 prevention. 43 [2551st words} 44 Suicide rates were analyzed for people 210 years only, as attributions of suicidal intent in younger children are 45 variable Age?specific suicide counts were tabulated based on National 1v?ital Statistics System coded death 4s certificate records (international Ciassrficotion of Diseases 10"? Revision, underlying-cause?of death codes X60- 4? X34, TRIO, Age-speci?c population estimates were obtained from us. Census Bureaufii-lational Center for 4-8 Health Statistics bridged?race population data releases. 50 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 5] from 1999-2016. Rate estimates were age-adjusted to the U.S. year 2000 standard population and expressed per 52 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data 53 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 54 are reported in terms of average annual percentage changes 55 56 Characteristics {Table 2) and circumstances {Table 3) of suicide decedents 210 years, with and without known 5? MHP, were compared in the 2? states with complete data participating in Nationai Violent Death 58 Reporting System [an?sl in 2015. defines MHP as disorders listed in the Diagnostic and Statistical 59 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 6 death certificates, coronerfmedical examiner reports {including toxicology), and law enforcement 62 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression to analyses estimated adjusted odds. ratios with 05% confidence intervals (CI), controlling for age group, sex, and 64 race! ethnicity. 65 RESULTS [5841600 words} as The most recent overall suicide rates [representing 2014?2016] varied four?fold, from 5.9 to 29.2 6? {Montana} per 100,000 persons per year {Table 1). Across the study period, rates increased in all states, except 68 Nevada {with the highest current suicide rate], with absolute increases ranging 69 (Wyoming) per 100,000. Percentage increases in rates ranged from +53% (Delaware) to (North Dakota), it) with increases of at least 30% observed in 25 states. Tl 1'2 7?3 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. T4 Nationally, the model-estimated AAPC for the overall suicide rate was By sex, the national MPC was 75 +11% for males and +15% for females. 76 Suicide decedents without known MHP iN=11,0391 were compared to those with MHP While all T8 decedents were predominately male (Table 2; 75.8%} and non-Hispanic white those without known if? MHP, relative to those with MHP, were more likely male [33.6% vs. 68.8%; adjusted odds ratio {awn-=23, 95% CI 80 2.2?2.5] and racial/ethnic minorities (odds ratio range: CI range Suicide 8 decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide iaDR 2.0, 82 95% CI of firearm suicide iaDR 1.6, 95% CI and oftesting positive for alcohol 1.2, 33 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever 34 served in the USE-Miners} 2 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall {Table 1' Comment Imam]: The comment made was that it wasn't clear how we defined MHP since substance use disorders are a diagnosis in Changed this to be more strai htl?orward. Comment [2319]: Scott we had this as 0.2 before?i' IEan you confirm 0.0 is n. correct? i . 3 Comment ham: Malia wanted to know i if we could say "had served or were i i serving" in the U.S. Military. i i i Flev 3.23.18 v3.1 '85 Although firearms were used most often, overall decedents with known MHP died by poisoning as significantly more than those without known [19.3% vs. most frequently by over-the- tf? counterfotherwise unclassified klrugs {35.8%l, opioids antidepressants or benzodiazepines as as All suicia's'aasasiig'ais know-d-MHF' and 9U circumstances information (Table 3). People without known MHP were 30% less likely to have any substance 9 abuse problems 95% While twovthirds of those with known MHP had a history of mental 92 health or substance abuse treatment just over half were in current treatment. 93 [Decedents without known MHP, versus those with known MHP had a significantly greater likelihood of any 94 relationship problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. 95 [115% vs. and recently perpetrating interpersonal violence vs. They ?nd were also significantly more likely to have experienced other life stressors, sueh as criminal-legal problems 97 vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 98 the preceding or upcoming two weeks [32.9% vs. Among both groups, the most common crises were 99 intimate partner [35.2% vs. 34.9%) and physical health problems [13.8% vs. respectively.] I ?1 Comment I think the comparison group 96?s are needed. For instance, drugs (35.8% vs. I think the presentation of the 93's would be easier here and later on in the results if the p?s came out and the text could simply reflect ?significant differences" or some variation of indicating in the groups were different. CHANGE THIS BIC lT OPENS UP THE ISSUE OF PEOPLE MHP TAKING antigens Comment lzn?]: I kept this paragraph con?ned to percentages but did add some Ladditional aOR's in other places. Comment lately]: We were asked to include disclosure of suicide intent. Also, 1 IOU Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among IUI those who were recently released, those without a known MHP were significantly more likely to be released l02 from a correctional facility {25.7% vs. or hospital [43.1% vs. 33.0%] than those with a known MHP. Among IflE decedents with known MHP recently released from an institution 42.3% were released from I04 facilities. 5 l05 [Decedents without known MHP, compared to those with known MHP, were 60% less likely to have a history of suicidal ideation 95% and "30% less likely to have an attempt history 95% 0.3). Both groups disclosed suicide intent frequently, {22.4% vs. respectivelyConclusions and Comments [68017130 words] The rise in suicide rates in the overall US. has been observed for many years Reporting of state-specific I It) trends overall and by sex is much less common. Similarly, while geographic patterning in suicide rates have II reportedtlcil, the current findings point to a disturbing pattern of increases nationwide. H2 Understanding the contributing circumstances of suicide is obligatory for prevention practice and decision- making. I I4 Research and prevention practitioners regularly state that suicide is not caused by a single factor, however, the I IS focus of suicide research and prevention practices almost solely focus on MHP. The current study found that I [6 more than half of suicide decedents in did not have a known This group suffered more relationship I 1? problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. I IS This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal If) deliberation IZU Among people with l'v'lHP, two?thirds had a history of mental health andfor substance abuse treatment and over IZI half were in current treatment. This suggests that additional supports for this population are needed to keep [22 them safe. This includes broader implementation of affordable and effective treatment modalities such as 23 doctor?patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to 24 behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems "H-u I I I kchanged the formatting here a bit. i {Comment wisqars I ment [1:119]: national strategy II I i Rev 3.23.18 v3.1 25 needed that Integrate physical and behavioral health and that better support suicide prevention and patient I26 safety, especially through care transitions Study findings indicate that people with known also experienced other life stressors such as jobifinancial, IZR relationship, andfor physical health problems. These findings point to the need to both prevent the conditions [29 associated with mental health problems in the first place and the need to support people with known IleP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities 132 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I33 economic supports ie.g., housing stabilization policies, household financial support}; teaching coping and I34 problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in [35 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, lilo emotional, and social support, and identifying and better supporting people at risk. Other strategies include l3? creating protective environments leg, reducing access to lethal means among people at risk, creating I38 organizational and workplace policies to promote help?seeking, easing transitions into and out of work for I39 people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe I40 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning and implementing such a comprehensive approach to suicide prevention I42 These findings have at least three limitations. In the state-level analysis, rankings for four states D, MA, RI, I43 might have been impacted by large proportions of injury deaths of undetermined Intent Ii.e. decreasing I44 suicide rates], or decreased percentages of such deaths overtime increasing suicide rates}. Second, I45 is not yet nationally representative, the 2? states included in the current study represent E41696 of the US. knowle-dge can impact data completeness and accuracy. Studies including in? I46 lpopulation. Third, abstractors of data are limited to information contained in investigative reports. I48 depth interviews with next-of-kin often see greater attributions to substance abuse disordereiuunun I49I however many methodological variations across studies exist It is likely that some people without known I50 MHF in the current study were experiencing mental health challenges that were unknown, and hence I52 importance of addressing the range of contributing circumstances. [53 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are I54 availablelStates and communities can use data from and resources such as CDC's Preventing Suicide: ISS Technical Package of Policies, Programs, and Practices [11] to better understand their suicide problem and I56 prioritize evidence?based comprehensive suicide prevention. Acknowledgments I53 The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I59 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. I60 Con?ict of Interest No conflicts of interest were reported. I62 I63 Corresponding author: Deborah M. Stone, dstone3@cdc.gov Jim-4383942 I64 I65 Author Affiliations: loo i?ivision of Violence Prevention, National Center for Injury Prevention and Control, 2Division of Analysis, [67 Research, and Practice Integration. National Center for injury Prevention and Control, CDC 4 Comment Is this confirmedJom i) I {Comment [1:119]: Need to consider this ll part further. llIComment Infill: Is this still ok? 1, Comment ham]: Cory had a great Idea here so took out, ental illness is an important risk factor for suicide, and is one of many requiring preventive action. She said the media would probablyjust take away Mental illness is an important risk factor, period. Rev 3.23.18 v3.1 68 References: I69 [1'0 Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March .15, 20.18. 2016. 2. Way?Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, IT3 Race/Ethnicity, Age Group, and Mechanism of Death United States, 200142015. MMWR Su Neill Summ, 1T4 2017. 66(18): p. 1-16. ITS 3. Kegier, S.R., OM. Stone, and KM. Holland, Trends in suicide by Ievei ofurbanization?United States, He 1999?2015. MMWR. Morbidity and mortality weekly report, 201?. 66(10): 270. 4. Kochanek, IL, et ah, Martaiity in the United States, 2015. NCHS Data Brief, no 293. National Center for ITS Health Statistics, 201?. I29 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai I 80 strategy for suicide prevention: goais and objectives for action: a report of the US Surgeon Generai and of the Nationai Action Suicide Prevention. 2012. IRQ 6. Rosen ma n, S.J., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. 169(2): p. 100?102. Iii-4 T. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide IBS Rate. 201?: Washington, D.C. I SE- 3. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicai [Sir Examiners. Archives of Suicide Research, 2010. 14(1): p. 2464. I 83 9. American Association, Diagnostic and statisticai manuai of mentai disorders 2013: I39 American Pub. 00 10. Li, 2., et al., Attributable risk of and soda-economic factors for suicide from individuai-ievei, popuiatian-hased studies: A systematic review. Social Science BI Medicine, 2011. 72(4): p. 503-616. 192 11. Stone, D.M., et ai., Preventing suicide: A technicai package ofpoiicies, programs, and practice. 2017. 193 12. Simon, T.R., et al., Characteristics afimpuisive suicide attempts and attempters. Suicide and Life? 194 Threatening Behavior, 2002. 32(51): p. 49-59. 195 13. World Health Organization, Risks to mentai heaith: An overview of and risk factors. [90 Geneva: WHO, 2012. 19? 14. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. [98 Injury prevention, 201?: p. injuryprev-2017?042366. I99 15. Hawton, K., et al., The autopsy approach to studying suicide: 0 review of methodaiogicai 200 issues.J Affect Disord, 1993. a. 269-?6. 2'01 Centers for Disease Control and Prevention, Web?based injury Statistics Query and Reporting System 202 Table 1 and Figure 1.doc 203- Tables 2 and 3.pdf 204 Word Count: 12631800 Comment lzu?il: Need to replace ref 10 and fix 11-15 3? 39 Flat.r 3.23.18 V3.1 Short title: Vital Signs: increasing Trends in State Suicide Rates and Fontributing Circumstances among people 210 years Deborah EE?kEtiie?riif?iifi'oiviaf hittifk's'c'cii?if E?ining Yuan, l'V'lS:1 Kristin M. Holland, Ashe Z. Ivey-Stephenson, Phi);1 Alex E. Crosby, M01 Structured abstract {2501'250 words?this word count is not included in the 1800 max for the remainder] Background: Suicide rates among people 210 years have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among people 210 years, by state and sex, across six consecutive three-year periods (1999-2016), were assessed using the National Vital Statistics System in 50 states and Washington, D.C Data from the National Violent Death Reporting in 20_15, examined precipitating circumstances among decedents with and without known MHP. Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of 30% or more. Rates increased significantly among males and females, in 34 and 43 states, respectively. Over half inf people did not have a known 46.0% dii?oppsexperienced similar and differing Comment Sometimes itls circumstances and some circumstances were common to both. Among people without known MHP, compared to those with 45.1% and 39.5% lp<.01ll had any relationship problemsfloss, 54.2% and 49.2% (p501) had any life stressorslloss and 32.9% and 26.0% (p501) had any recentlimpending crises, respectively. Conclusions: Suicide rates increased significantly across most states from 1999-2016. Various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {250f250 words} in 2016, nearly 45,000 suicides l15.6f100,000l occurred in the United States among people 210 years old Between 1999 and 2015, suicide rates increased sexes, racial/ethnic groups, and urbanization levels 31. Suicide is the 10Lh leading cause of death and is one of just three leading causes to be increasing Additionally, rates of Emergency Department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs The Notional Strategy for Suicide Preventionj'NSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels li.e., individual, family/relationship, community, and societalj. Such an approach underscores that suicide is rarely caused by any single factor alone, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing factors include social and economic problems, access to lethal means substances, firearms, bridges] among people at risk, poor coping and problem-solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from can help reach the nation?s goal of reducing suicide rates 20% by 2025 To assist states I [Comment Izal?Dj: Is this stillaccurate? referred to as a 'contributing circumstance? and sometimes its referred to as a ?contributing factor?I . i wonder if its best to be consistent so as not to confuse the reader that the two are different. From a plain language perspective, I prefer contributing factors Also, sometimes it is referred to a just ?contributing factor' while other times it is referred to as ?contributing risk factor?. Also, sometime it is referred to as *contributing? and sometimes as *assoclated'. Should we be consistent? I FOUND: PRECIPITATING CIRCUMSTANCES IALONEFE FACTORS RISK 1 Comment In?ll: Switched the order Lhere Rey 3.23.18 v3.1 40 in achieving this goal, this study analyzes statevspecific trends in suicide rates, assesses the multiple contributing 4 factors, and provides recommendations for multi-level comprehensive suicide prevention. METHODS l255l250 words} 43 Suicide rates were analyzed for people 210 years only, as attributions of suicidal intent in younger children are 44 variable Age-specific suicide counts were tabulated based on National Vital Statistics System coded death 45 certificate records [international Ciossificotion of Diseases 10?" Revision, underlying-cause-of death codes 46 X34, Age-specific population estimates were obtained from 0.5. Census Bureaui?National Center for 47 Health Statistics bridged?race population data releases. 48 49 National and state?level suicide rate estimates were calculated for six consecutive three?year aggregate periods 50 from 1999-2015. Rate estimates were age?adjusted to the 0.5. year 2000 standard population and expressed per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same three-year data 5] aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 52 53 are reported in terms of average annual percentage changes 54 55 Characteristics [Table 2) and circumstances {Table 3} of suicide decedents 210 years, with and without known 56 MHF, were compared in the 27 states with complete data participating in National Violent Death 5? Reporting System in 2015. defines MHP as disorders listed in the Diagnostic and Statistical 58 Man ual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 60 death certificates, coronerfmedical examiner reports {including toxicology), and law enforcement 6 reports. Decedents with and without known MHP Iwere compared Using Chi-Square tests. Logistic regression 62 analyses estimated adjusted odds ratios with 95% con?dence intervals controlling for age group, sex, and 59 defined as MHP, and were examined in both aggregates 3993355 63 racer'ethnicity. 64 RESULTS [584!600 words] 65 The most recent ovarall suicide rates [representing 2014-2015] varied four-fold, from 5.9 to 29.2 66 (Montana) per 100,000 persons per year {Table Across the study period, rates increased in all states. except 67 Nevada (with the highest current suicide rate), with absolute increases ranging fromli-DB {Delaware?to?gl {38 {Wyoming} per 100,000. Percentage increases in rates ranged from +5.51% [Delaware] to 616% (North Dakota) with increases of at least 30% observed in 25 states. T0 il Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall (Table ?2 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. ?i3 Nationally, the modelaestimated for the overall suicide rate was By sex, the national AAPC was 74 +11% for males and +15% for females. i5 T6 Suicide decedents without known MHP iN:11,039} were compared to those with While all 77 decedents were predominately male (Table 2; 15.8%} and non?Hispanic white those without known Y8 MHP, relative to those with MHP, were more likely male [33.6% vs. adjusted odds ratio 95% CI i9 2.2-2.5) and racial/ethnic minorities (odds ratio range: 1.2-2.1; 95% CI range Suicide 80 decedents without known MHP also had significantly greater odds of perpetrating homicide?suicide 2.9, til 95% CI of firearm suicide 1.6. 95% CI 151.7}, and of testing positive for alcohol iaOR 1.2, 82 95% Cl Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, ever ., Comment Imam]: The comment made was that it wasn't clear how we defined MHP since substance use disorders are a diagnosis in Changed this to be more strai htforward. i Comment [1319}: Scott we had this as 0.2 before?i' l?an you confirm 0.0 is correct? I Cornment za?ilz Malia wanted to know If we could say "had served or were serving" in the U.5. Military. ~quserved in the 0.5. ['nilitan} Flev 3.23.18 v3.1 84 Although firearms were used most often, overall decedents with known MHP died by poisoning 85 significantly more than those without known [19.3% vs. most frequently by over-the- 86 counterfotherwise unclassified brugs opioids antidepressants or benzodiazepines a? as All suica's'aa'csasas'a?a know-n-MHF' (wagon and 39 circumstances information (Table People without known MHP were 30% less likely to have any substance 9D abuse problems 95% While twovthirds of those with known MHP had a history of mental 9i health or substance abuse treatment just over half were in current treatment. 92 [Decedents without known MHP, versus those with known MHP had a significantly greater likelihood of any 93 relationship problemfloss [45.1% vs. specifically intimate partner problems [30.2% vs. 94 [115% vs. and recently perpetrating interpersonal violence vs. They 95 Iwere also significantly more likely to have experienced other life stressors, soch as criminal-legal problems 96 vs. or evictionfloss of home vs. and they were more likely to have had a crisis within 97' the preceding or upcoming two weeks [32.9% vs. Among both groups, the most common crises were 98 intimate partner [35.2% vs. 34.9%) and physical health problems [13.8% vs. respectively.] . .4 99 Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among lliU those who were recently released, those without a known MHP were significantly more likely to be released from a correctional facility {25.7% vs. or hospital [43.7% vs. 33.0%] than those with a known MHP. Among decedents with known MHP recently released from an institution 42.3% were released from IDS facilities. l04 [Decedents without known MHP, compared to those with known MHP, were 60% less likely to have a history of IBIS suicidal ideation 95% and "30% less likely to have an attempt history 95% lilo Both groups disclosed suicide intent frequently, {22.4% vs. respectively] Conclusions and Comments words] lflli The rise in suicide rates in the overall U.5. has been observed for many years Reporting of state-specific [09 trends overall and by sex is much less common. Similarly, while geographic patterning in suicide rates have I ll) frequently been reported the current findings point to a disturbing pattern of increases nationwide. II Understanding the contributing circumstances of suicide is obligatory for prevention practice and decision- l2 making. IIB Research and prevention practitioners regularly state that suicide is not caused by a single factor, however, the I I4 focus of suicide research and prevention practices almost solely focus on MHP. The current study found that I [5 more than half ofsuicide decedents in did not have a known This group suffered more relationship llo problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises I This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Ill} Among people with l'v'lHP, two?thirds had a history of mental health andfor substance abuse treatment and over l2ll half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as 22 doctor?patient collaborative care models and cognitive?behavioral therapy. Additionally, greater access to 23 behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems Commont I think the comparison group 96?s are needed. For instance, drugs (35.8% vs. I think the presentation of the would be easier here and later on in the results if the p's came out and the text could simply reflect ?significant differences? or some variation of indicating in the groups were different. CHANGE THIS BIC lT OPENS UP THE OF PEOPLE MHP TAKING tillT'PEi 51C 3 Comment lentil]: I kept this paragraph con?ned to percentages but did add some Ladditional aOR's in other places. Comment lately]: We were asked to include disclosure of suicide intent. Also, 1 ,changed the formatting here a bit. Flev 3.23.18 v3.1 I24 needed that integrate physical and behavioral health and that better support suicide prevention and patient I25 safety, especially through care transitions I26 Study findings indicate that people with known IleP also experienced other life stressors such as jobi?flnancial, I2?r relationship, andior physical health problems. These findings point to the need to both prevent the conditions [28 associated with mental health problems in the first place and the need to support people with known MHP to [29 decrease their risk of poor social, health, and economic outcomes I30 These results, together, underscore the importance of comprehensive statewide suicide prevention activities that address multiple factors associated with suicide. Prevention strategies may include: strengthening I32 economic supports leg, housing stabilization policies, household financial support}; teaching coping and I33 problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in [34 life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, I35 emotional, and social support, and identifying and better supporting people at risk. Other strategies include 36 creating protective environments le.g., reducing access to lethal means among people at risk, creating I37 organizational and workplace policies to promote help?seeking, easing transitions into and out of work for I33 people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe 39 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning I40 and implementing such a comprehensive approach to suicide prevention I4I These findings have at least three limitations. In the state-level analysis, rankings for four states D, MA, RI, I42 might have been impacted by large proportions of injury deaths of undetermined Intent ii.e. decreasing I43 suicide rates], or decreased percentages of such deaths overtime increasing suicide rates}. Second, I44 is not yet nationally representative, the 2? states included in the current study represent E41696 of the US. I46 There-f-drefthe extent of infofma-rit- knowle-dge can impact data completeness and accuracy. Studies including in? I45 Population. Third, abstractors of data are limited to information contained in investigative reports. depth interviews with next-of-kin often see greater attributions to substance abuse disordereiuunun I48 however many methodological variations across studies exist It is likely that some people without known I49 MHF in the current study were experiencing mental health challenges that were unknown, and hence I 5 I importance of addressing the range of contributing circumstances. [52 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are I53 availablelStates and communities can use data from and resources such as CDC's Preventing Suicide: . l54 Technical Package of Policies, Programs, and Practices [11] to better understand their suicide problem and I55 prioritize evidence?based comprehensive suicide prevention. I5o Acknowledgments The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I58 Statistics, National Center for Health Statistics, CDC, for their statistical consultation. 59 I?fl Con?ict of Interest No conflicts of interest were reported. [62 Corresponding author: Deborah M. Stone, dstone3@cdc.gov noses-3942 I63 I64 Author Affiliations: l?ivision of Violence Prevention, National Center for Injury Prevention and Control, 2Envision of Analysis, [66 Research, and Practice Integration. National Center for injury Prevention and Control, CDC 4 Comment Is this confirmedJom P) {Comment [1919]: Need to consider this it part further. llIComment Infill: Is this still ok? 1, Comment pars]: Cory had a great idea here so took out, ental illness is an important risk factor for suicide, and is one of many requiring preventive action. She said the media would probablyjust take away Mental illness is an important risk factor, period. Rev 3.23.18 v3.1 I??i References: [08 169 (WISCIARSJ. Atianta, GA: Nationai Center for injury Prevention and Controi. Retrieved March .15, 20.18. ITU 2016. 2. Way?Stephenson, A.Z., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, [12 Race/Ethnicity, Age Group, and Mechanism of Death United States, 200142015. MMWR Su Neill Summ, I?i3 2017. 66(18): p. 1-16. I?i4 3. Kegier, S.R., OM. Stone, and KM. Holland, Trends in suicide by ievei ofurbanization?United States, ITS 1999?2015. MMWR. Morbidity and mortality weekly report, 201?. 66(10): 270. l'i6 4. Kochanek, IL, et ah, Martaiity in the United States, 2015. NCHS Data Brief, no 293. National Center for Health Statistics, 2012. ITS 5. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goats and objectives for action: a report of the US Surgeon Generai and [80 of the Nationai Action Suicide Prevention. 2012. I 8 6. Rosen ma n, S.J., Preventing suicide: what work and what not. The Medical Journal of Australia, 82 1998. 169(2): p. 100?102. 7. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide 84 Rate. 201?: Washington, D.C. 185 3. Crepeau-Hobson, F., The Autopsy and Determination of Chiid Suicides: A Survey of Medicai ISIS Examiners. Archives of Suicide Research, 2010. 14(1): p. 2464. 8? 9. American Association, Diagnostic and statisticai manuai of mentai disorders 2013: 138 American Pub. IRS 10. Li, 2., et al., Attributable risk of and soda-economic factors for suicide from individuai-ievei, 90 popuiatian-hased studies: A systematic review. Social Science BI Medicine, 2011. 72(4): p. 603-616. l9 11. Stone, et ai., Preventing suicide: A technicai package ofpoiicies, programs, and practice. 2017. [92 12. Simon, T.R., et al., Characteristics afimpuisive suicide attempts and attempters. Suicide and Life? 193 Threatening Behavior, 2002. 32(51): p. 49-59. 194 13. World Health Organization, Risks to mentai heaith: An overview of and risk factors. [95 Geneva: WHO, 2012. 96 14. Caine, E.D., et al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. 19? Injury prevention, 201?: p. injuryprev-2017?042366. I93 15. Hawton, K., et al., The autopsy approach to studying suicide: 0 review of methodaiogicai issues.J Affect Disord, 1993. a. 269-26. 200 Centers for Disease Control and Prevention, Web?based injury Statistics Query and Reporting System 201 Table 1 and Figure 1.doc 202 Tables 2 and 3.pdf 203 Word Count: 1262,!1800 Comment lzu?ll: Need to replace ref 10 and fix 11-15 I Rev 3.23.18 v3.1 Short tItle VItal Signs increasing Trends in State Suicide Rates and Fontributing Circumstances among people 2 210 years 3 Deborah M'S-tone, Thomas Rr?Sfm-on P'hlIi;1 Katherine A. Fowler, Scott R. Kegler, 2 Kemlng Yuan, 4 Kristin M. Holland, Ashe Z. Ivey-Stephenson, PhD;1Alex E. Crosby, M01 5 Structured abstract [2501'250 words?this word count is not included in the 1800 max for the remainder] a Background Suitide rates among people 210 years have risen nearly 30% since 1999 Mental health problems 7 are Just one factor contributing to suicide. Examining state-level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning i 9 Methods Trends In age adjusted suicide rates among people 210 years, by state and sex. across six consecutive If) three- -year periods (1999- -2016l, were assessed using the National Vital Statistics System in 50 states and I Washington D. (D (2..) Data from the National Violent Death Reporting System hovering}? states in 2015, a i [2 u_sed to examined contributing among decedents with and witho?utltnown MHP. Results From 1999 2016, suicide rates increased significantly in 44 states, with 25 states experiencmg Increases l4 of 30% or more. Rates Increased significantly among males and females, in 34 and 43 states respectively Over [5 half [54. bf decedentspeoele did not have a known MHP, Among decedents with circumstance lo Informatior?l those without known MHP. were more lilcelv lall as .01} 9 stressorsiloss 2% vs and?49 and 20 recent/impending crises {32. 9% vs 25. (Ml, but these circumstances were common across groups. ad?aay?l-ife 21 . . . - - 7'9 Conclusnans rates Increased significantly across most states from 1999-2016 Various Circumstances 23 contributed to suicides among people with and without known MHP lsthisstillaccurate? Comment Sometimes it is referred to as a rcontributing circumstance? and sometimes its referred to as a 'contributing factor?. i wonder if its best to be consistent so as not to confuse the reader that the two are different. From a plain language perspective, I prefer contributing factors i2). Also, sometimes it is referred to a just ?contributing factor' while other times it is referred to as ?contributing risk factor?. Also, sometime it is referred to as and sometimes as ?associated'. Should we be consistent? I FOUND: CONTRIBUTING PRECIPITATING CIRCUMSTANCES CONTRIBUTING FACTORS IALDN El=3 .ul'l. Comment lthink we can say contributing circumstances a few times in the text and then just use circumstances Comment [tgs9 : Some editing is needed because this read like the data examined circumstances. rComment no Switched the order 1 24 Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach I, here I I h. 25 to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help 1 . . . . . Comment ltgs9]: I saw that Cory made a to after a suicide occws. 2? INTRODUCTION 28 BACKGROUND AND PURPOSE {2501'250 words} I 3] the 10Lh IeadIng cause of death and is one of just three leading causes that arete?be Increasing 32 [1 4] 33 Increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more 34 than $59 billion in direct medical and work loss costs 35 The Notional Strotegyfor Suicide Preventionj'NSSP} calls for a public health approach to suicide prevention 36 with efforts spanning across multiple levels li.e., individual, familyfrelationship, community, and societal} Such 3? an approach underscores that suicide is rarely caused by any single factor?alone, but rather, is multi determined 33 [I'riqvbe su_b_sL1tute good suggestion to say among those with circumstance information. If we make one point about the p-value we can save words. The edits I suggested allow us that ?common for both. Comment Are we going to have to explain this such as putting in {rate L15.0 per 100,000 population} Comment ICAI: Alternate citation though not quite covering the same - Curtin SC, Warner M, Hedegaard H. Suicide rates forfemales and males by race and ethnicity: United States, 1999 and 2014. NCHS Health E?Stat. National here are differences but these are rru tiisrc Ied_j Despite the NSSP guidance, suicide prevention efforts largely focus on 39 and treating individuals with mental health problems Other contributing faetoes 1 gCenter for Health Statistics. April 2016. Flev 3.23.18 v3.1 Comment wars]: The comment made was that It wasn't clear how we defined MHF since substance use disorders are a diagnosis in Changed this to be more strai htforward. Comment [2319]: Scott we had this as 0.2 before?i' Can you confirm 0.3 is correct? 40 circumstances include social and economic problems, access to lethal means substances, firearms, bridges) 4 among people at risk, poor coping and problem?solving skills, and prior suicide attempts Expanded 42 awareness of the additional circumstances that contribute to suicide risk apart from MHP, can help reach the i 43 nation?s goal of reducing suicide rates 20% bv 2025 To assist states in achieving this goal, this studv analvzes i 44 statesspecific trends in suicide rates. assesses the multiple contributing factors, and provides recommendations i 45 for multi-Ievel comprehensive suicide prevention. 46 METHODS [255/250 words} 5 47 Suicide rates were analyzed for people 210 years onlv, as attributions of suicidal intent in vounger children are ,i 48 variable Age-specific suicide counts were tabulated based on National lv?ital Statistics System coded death i 49 certificate records [international Ciossiificotion of Diseases 10'? Revision, underlying?cause?of death codes X34, Age-specific population estimates were obtained from U.5. Census Bureaui'National Center for SI Health Statistics bridged-race population data releases. 52 53 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 54 from 1559-2015. Fiate estimates were age-adjusted to the US. year 2000 standard population and expressed per 55 100,000 persons per vear. Age?adjusted suicide rate trends were modeled using the same three-vear data 56 aggregates, emploving weighted least squares regression with inverse-variance weighting. Modeled rate trends 5? are reported in terms of average annual percentage changes 58 59 Characteristics [Table 2] and cirCumstances {Table 3} of suicide decedents 210 years, with and without known 60 MHF, were compared in the 2? states with complete data participating in National Violent Death 6 Reporting Svstem in 2015. defines MHP as disorders listed In the Diagnostic and Statistical 62 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 63 defined as MHP, and were examined in both groupsbi'v?DRS aggregates data from three primarv data sources: 64 death certificates, coronerfmedical examiner reports I reports. Decedents with and without known MHP were compared using Chi-square tests. Logistic regression as analvses estimated adjusted odds ratios with 95% confidence intervals controlling for age group. sexrace/ethnicity. as RESULTS (sea/sou wordsi 69 The most recant overall suicide rates (representing 2014?2015] varied four?fold, from 5.5 to 25.2 20 (Montana) per 100,000 persons per vear (Table Across the studv period, rates increased in all states, except Nevada {with the highest current suicide rate), with absolute increases ranging from 0.8 22 {Wyoming} per 100,000. Percentage increases in rates ranged from +55% (Delaware) to 62.6% {North Dakota), 23 with increases of at least 30% observed in 25 states. 24 i5 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the US. overall {Table 1] 76 By sex, modeled rate trends indicated signi?cant increases in 34 states for males and in 413 states for females. Nationally, the modelvestimated MPC for the overall suicide rate was Ev sex, the national MPC was 7?8 use; for males and +16% for females. 79 80 Suicide decedents without knowo MHP {hi?111,039} were compared to those with While all til decedents were predominatelv male (Table 2: 75.8%} and non?Hispanic white those without known 82 MHP, relative to those with MHP, were more likelv male [33.5% vs. 53.8%; adjusted odds ratio iaORi=2.3, 95% Cl '83 2.2?2.5] and racial/ethnic minorities (odds ratio range: 1.2-2.1; 95% Cl range - Suicide B4 decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide iaDFi 2.9, HS 55% CI of firearm suicide 1.5, 95% CI and of testing positive for alcohol 1.2, 2 91s 98 99 100 IOI Ill? MB [05 we Ill? IDS 112 113 114 115 117 113 119 120 121 122 123 IM- l25 [Eff Rev 3.23.18 v3.1 95% CI Among adult decedents, and 15.3% of people without and with MHP, respectively, ever served in the U.S. ilitary} Although firearms were used most often, overall decedents with known MHP were nlqgejj?ygdied by suffocation (31.3 vs. 26.9%l and poisoning 119.8% vs. 111.4%] signifieantiy?mere than those without known MHP . . . .F All suicide decedents with known MHP and approximately 35% without MHP had available circumstances information {Table 3). People without known MHP were 39% ess likely to have any substance abuse problems 95% While two-thirds of those with known MHP had a history of mental health or substance abuse treatment just over half were in current treatment. @ecedents without known MHP, versus those with known lv'lHP had a significantly greater likelihood of any relationship problemi'loss (45.1% vs. specifically intimate partner problems [30.2% vs. {115% vs. and recently perpetrating interpersonal violence vs. They were also significantly more likely to have experienced other life stressors, such as criminal?legal problems (10.3% vs. or eviction/loss of home vs. and they were more likely to have had a crisis within the preteding or upcoming two weeks [32.9% vs. Among both groups, the most common crises were intimate partner [36.2% vs. 34.9%} and physical health problems [13.3% vs. respectively] Decedents without known MHP had signi?cantly lower odds of recent release from any institution, but among those who were recently released l5'llt?3l1 those without a known MHP were significantly more likely to be released from a correctional facility 1:7? area's)? 'a'if? MHP. Among decedents with known lv'lHP recently released from an institution 42.3% were released from facilities. @ecedents without known MHP, compared to those with knewn?MHF', were Ease-less likely to have a history of suicidal ideation 95% and laid?less?ii-kely?ta?have?an? prior suicide attempt histew 95% Both groups disclosed suicide intent frequently, [22.4% vs. respectively] Conclusions and Comments [680/700 words] The rise in suicide rates in the overall U.S. has been observed for many years [Reporting of state?specific trends were? and sex is mush '655 . frequently been reported the current findings point to a disturbing pattern of increases nationwide. Understanding the contributing circumstances of suicide is obligatory in Iayhe s- 1115: note "es-stirInaI?) for prevention practice and decision-making. Researchg and pseventien?practitioners regularly state that suicide is not caused by a single factor, however, the focus ofsuicide research and prevention practices almost solely focus on MHP. The current study found that more than half of suicide decedents in did not have a known MHP. This group suffered more relationship problems and life stressors such as criminal-legal matters, evictionfioss of home, and recent or impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts occur with minimal deliberation Among people with MHP, two?thirds had a history of mental health andfor substance abuse treatment and over half were in current treatment. This suggests that additional supports for this population are needed to keep them safe. This includes broader implementation of affordable and effective treatment modalities such as doctor?patiEnt collaborative care models and cognitive?behavioral therapy. Additionally, greater access to 3 Comment IFCH: lthink the comparison Comment lentil]: We were asked to 1 Comment lately]: Malia wanted to know if we could say "had served or were serving? in the Military. [Comment ltgsii}: I think that is fine. group are needed. For instance, drugs [35.3% vs. I think the presentation of the would be easier here and later on in the results if the p's came out and the text could simply reflect ?significant differences? or some variation of indicating in the groups were different. CHANGE BIC IT OPENS UP THE ISSUE OF PEOPLE WED MHP TAKING ANTIDEP ETC Do we eyen need to give this distribution? We don?t talk about the types in the discussion .a Comment [21:19]: I kept this paragraph con?ned to percentages but did add some additional aC-R's in other places. 1 Comment tgs9l: We should be consistent and add this if we have the below. J5. include disclosure of suicide intent. Also, Lchanged the formatting here a bit. Comment [tg59]: Why did you add this? it sounds like others have done it already. I think the last version of the start of the Lconcltoion Was more compelling. Rev 3.23.18 v3.1 i[Com me nt 1mm: Is this confirmedLTom - ., risk factor, period. behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems I28 needed that into rate sical and beh v'oral alth and that better su rt su?c'de eventComment [tgs9]: This is from the 2015 I29 safety, especially through care transmons NUDRS 55 I30 Study findings indicate that people with known MHP also experienced other life stressors such as jobffinancial, EIComment ICAI: Good addition I3l relationship, andfor physical health problems. These findings point to the need to both prevent the conditions Comment In?ll: Need to consider this 1' I32 associated with mental health problems in the first place and the need to support people with known to part furthen I33 decrease the? risk of poor SDCIBI, health, and economlc outcomes IffIC?mem In?ll: Is this Still ok? I34 These results, together, underscore the importance of comprehensive statewide suicide prevention activities Comment [1319]: Cary had a great Idea I35 that address multiple factors associated with suicide. Prevention strategies may include: strengthening [It'll herESO took out, ?nial Illness is an 36 economic supports housing stabilization policies, household ?nancial support}; teaching coping and ?fti imporgant ?5k factorfor ?dis problem?solving skills to manage everyday stressors and prevent future relationship problems, especially early in Ilili life: promoting social connectedness to increase a sense of belongingness and access to informational, tangible, {Elli take away Mental illness is an important I33 I39 emotional, and social support, and identifying and better supporting people at risk. Other strategies include I40 creating protective environments reducing access to lethal means among people at risk, creating I4I organizational and workplace policies to promote help-seeking, easing transitions into and out of work for .1 I42 people with lleP and other life challenges], supporting family and friends after a suicide, and assuring safe I43 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning I44 and implementing such a comprehensive approach to suicide prevention I45 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, I46 might have been impacted by large proportions of injury deaths of undetermined intent decreasing l4? suicide rates), or decreased percentages of such deaths over time He. increasing suicide rates}. Second, I48 is not yet nationally representative, the 2? states included in the current study represent[49?% of the U5. . I49 lpopulation. Third, abstractors of data are limited to information contained in investigative reports. l'r? I50 Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including in- I: I55 importance of addressing the Lora-.1 range of contributing circumstances. ,i I I56 Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available States and communities can use data from and resources such as CDC's Preventing Suicide: or . I58 Technical ?o-dkoEe-of?olicies, Programs, and Practices [11] to better understand their suicide problem and I59 prioritize evidence-based comprehensive suicide prevention. I60 Acknowledgments l6l The authors acknowledge Robert Anderson, Holly Hedegaard, and Margaret Warner from the Division of Vital I62 Statistics, National Center for Health Statistics. CDC, for their statistical consultation. [63 I64 Con?ict of Interest No conflicts of interest were reported. I65 lac Corresponding author: Deborah M. Stone, dagoneEchc-gov Fi?-4883942 I o? 68 Author M?liations: Rev 3.23.18 v3.1 I69 1Division of Violence Prevention, National Center for Injury Prevention and Control, aDivision of Analysis, 170 Research, and Practice Integration, National Center for injury Prevention and Control, CDC References1140 192 193 1:has!statistics-testesenasset Centers for Disease Control and Prevention, Web-based injuryr Statistics Query and Reporting System Ationto, GA: Nationai Center for injuryr Prevention and Controi. Retrieved March 15, 2018. 2016. Nev-Stephenson, A.2., et al., Suicide Trends Among and Within Urbanization Leveis by Sex, RacefEthnicity, Age Group, and Mechanism of Death United States, 2001?2015. MMWR Surveill Summ, 2017. 66(18): p. 1-16. Kegier, S.R., 0M. Stone, and KM. Holland, Trends in suicide by ievei of urbanization?United States, 1999-2015. MMWR. Morbidity and mortality weekly report, 201?. 66(10): 220. Kochanek, It, et ai., Mortaiity in the United States, 201? 6. NCHS Data Brief; no 293. National Center for Health Statistics, 2017. Office of the Surgeon General and National Action Alliance for Suicide Prevention, 2012 Nationai strategy for suicide prevention: goats and objectives for action: a report of the US Surgeon Generai and of the National Action for Suicide Prevention. 2012. Rosen ma n, 5.1., Preventing suicide: what work and what not. The Medical Journal of Australia, 1998. Torguson, K. and A. O'Brien, Leading Suicide Prevention Efforts Unite to Address Rising Nationai Suicide Rate. 201?: Washington, D.C. Crepeau?Hobsan, P, The Autopsy and Determination of Child Suicides: A Survey of Medicai Examiners. Archives of Suicide Resea rch, 2010. 14(1): p. 24?34. American Association, Diagnostic and statisticai manuai of mentai disorders 2013: American Li, 2., et al., Attributabie risk of and socia-economicfactors for suicide from individuai-ievei, popuiation-based studies: A systematic review. Social Science 81 Medicine, 2011. 22(4): p. 508-616. Stone, 0.90., et al., Preventing suicide: A technicai package of paiicies, programs, and practice. 2017. Simon, T.R., et al., Characteristics of impuisive suicide attempts and attempters. Suicide and Life- Threatening Behavior, 2002. 32(51): p. 49-59. World Health Organization, Risks to mentoi heaith: An overview of and risk factors. Geneva: WHO, 2012. Caine, 12.0., at al., Comprehensive, integrated approaches to suicide prevention: practicai guidance. Injury prevention, 201?: p. injuryprev?ZOlT-OAHEB. Hawtan, K., et al., The autopsy approach to studying suicide: a review of methodoiogicai issues. Affect Disord, 1998. p. 259-76. 205 Table 1 and Figure 1.doc 206 Tables 2 and 3.pdf 20? Word Count: 126210300 Comment lzu?ll: Need to replace ref 10 and fix 11-15 lad H-IGHUI Rev 3.25.18 v3.2 Short title: ncreasing Trends in State Suicide Rates and Contributing Circumstances among people all) years] Comment Need to edit this. Deborah M. Stone, Thomas R. Simon Ph0;? Katherine A. Fowler, Ph0;' Scott R. kegler, Keming Yuan, l'v'lii;JL Kristin M. Holland, Asha Z. Ivey-Stephenson, Ale): E. Crosby, M01 Structured abstract [2451250 words?this word count is not included in the 1800 max for the remainder] Background: Suicide rates have risen nearly 30% since 1999. Mental health problems are just one factor contributing to suicide. Examining state?level trends in suicide and other contributing circumstances can inform comprehensive state suicide prevention planning. fl Methods: Trends in age?adjusted suicide rates among people aged 210 years, by state and sex, across six consecutive three?year periods (1999?2016], were assessed using data from the National Vital Statistics System for 50 states and Washington, DC Data from the National Violent Death Reporting System, hoverindgEr'mJl states in 1015, were used to examine contributing circumstances among decedents with and without known MHP. Results: From 1999?2016, suicide rates increased significantly in 44 states, with 25 states experiencing increases of 30% or more. Rates increased signi?cantly among males and females, in 34 and 43 states, respectively. Over half of decedents did not have a known MHP. Among decedents with circumstance information, those without known MHP were more likely {all p501] than those with a MHP to have relationship problemsi?loss (45.1% vs life stressorsfloss {54.2% vs 49.7%] and recent/impending crises {32.9% vs but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states from 1999-2015. 1various circumstances contributed to suicides among people with and without known MHP. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support people at risk, prevent reattempts, and help friends/family after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE {255,4'250 words] In 2016, nearly 45,000 suicides [agaradjustedjl occurred in the United States among people 210 years old Between 1999 and 2016, suicide rates increased across sexes, racial/ethnic groups, and urbanization levels Suicide is the 10?? leading cause of death and is one of just three leading causes that are increasing Additionally, rates of Emergency Department visits for nonfatal self-harm,a key risk factor for suicide, increased nearly 45% betWeen 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than SEE-9 billion in direct medical and work loss costs The National Strategiffor Suicide PreventioniNSSP} calls for a public health approach to suicide prevention with efforts spanning across multiple levels individual, family/relationship, community, and societal]. Such an approach underscores that suicide is rarely caused by any single factor, but rather, is multi-determined. Despite the NSSP guidance, suicide prevention efforts largely focus on identifying and treating individuals with mental health problems Other contributing circumstances include social and economic problems, access to lethal means leg, substances, firearms, bridges] among people at risk, poor coping and problem- solving skills, and prior suicide attempts Expanded awareness of the additional circumstances that contribute to suicide risk apart from MHP, and action to address them, can help reach the nation?s goal of Could be interpreted that contributing circumstances are increasing as well. {[Comment [1319]: Is this still accurate? Rev 3.25.18 v3.2 40 reducing suicide rates 20% by 2025 To assist states in achieving this goal, this study analyzes state-specific trends in suicide rates, assesses the multiple contributing factors, and provides recommendations for multi?level 4 42 comprehensive suicide prevention. 43 METHODS (256,050 words} 44 Suicide rates were analyzed for people aged 210 years only, as attributions of suicidal intent in yoUnger children 45 are variable Age?specific suicide counts were tabulated based on National Vital Statistics System coded death 4s certi?cate records (international Ciassr?cotion of Diseases 10?? Revision, underlying-cause?of death codes X60- 4? X34, TRIO, Age-specific population estimates were obtained from 0.5. Census Bureauill-lational Center for 48 Health Statistics bridged?race population data releases. 49 50 National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods 5] spanning 1999-2015. Rate estimates were age-adjusted to the 0.5. year 2000 standard population and expressed 52 per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data 53 aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends 54 are reported in terms of average annual percentage changes 55 56 Characteristics {Table 2) and circumstances (Table 3) of suicide decedents 210 years, with and without known 57 MHP, were compared in the 2? states with complete data participating in National Violent Death 58 Reporting System in 2015. defines MHP as disorders listed in the Diagnostic and Statistical 59 Manual of Mental Disorders, Fifth Edition For this study, alcohol and other substance use disorders were not 6 death certificates, coronerfmedical examiner reports (including toxicology), and law enforcement 62 reports. Decedents with and without known MHP were compared using Chi?square tests. Logistic regression to analyses estimated adjusted odds ratios with 95% confidence intervals (CI), controlling for age group, sex, and 64 raceiethnicity. as assurrs (seafood words] as The most recent overall suicide rates (representing 2014?2016] varied four?fold, from 5.9 to 29.2 6? (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states, except 68 Nevada (with the 9?1 highest current suicide rate], with absolute increases ranging from +0.8 (Delaware) to +3.1 69 (Wyoming) per 100,000. Percentage increases in rates ranged from +99% (Delaware) to 67.6% (North Dakota), it) with increases more than 30% observed in 25 states. 'il ?2 Modeled suicide rate trends indicated significant increases for 44 states, as well as for the U5. overall (Table i3 By sex, modeled rate trends indicated significant increases in 34 states for males and in 43 states for females. T4 Nationally, the model-estimated AAPC for the overall suicide rate was By sex, the national AAPC was 75 +11% for males and +2.59; for females. 76 Suicide decedents without known MHP were compared to those with MHP While all is decedents were predominately male (Table 2; 75.8%} and non?Hispanic white those without known i9 MHP, relative to those with MHP, were more likeiy male (33.6% vs. 68.8%; adjusted odds ratio 95% CI 80 2.2?2.5] and racial/ethnic minorities (odds ratio range: Cl range Suicide 8 decedents without known MHP also had significantly greater odds of perpetrating homicide-suicide 2.9, 82 95% CI of firearm suicide 1.6, 95% CI and oftesting positive for alcohol 1.2, 33 95% CI Among adult decedents, 20.1% and 15.3% of people without and with MHP, respectively, [ever 34 served, or were serving,}n the 0.5. military. 2 Comment Inf"): The comment made was that it wasn't clear how we defined MHP since substance use disorders are a diagnosis in Changed this to be more strai htforward. [Comment ham]: Correct definition for military services? Rev 3.25.18 v3.2 85 Although firearms were used most often, overall decedents with known MHP were more likely to die by as suffocation (31.3 vs. 26.9%] and poisoning {19.8% vs. 10.4%] than those without knowo MHP . 3? All suicide decedents with known MHP iN=9,407} and approximately 35% without MHP had available 88 circumstances information {Table People without known MHP were less likely to have any substance abuse 89 problems 95% While two-thirds of those with known MHP had a history of mental health 90 or substance abuse treatment just over half were in current treatment. 9 [Decedents without known MHP, versus those with known lv'lHP had a significantly greater likelihood of any 92 relationship problemfloss [45.1% vs. specifically intimate partner problems {30.2% vs. 93 [115% vs. and recently perpetrating interpersonal violence vs. They 94 were also significantly more likely to have experienced other life streSSors, such as criminal-legal problems 95 vs. or eviction/loss of home vs. and they were more likely to have had a crisis within as the preceding or upcoming two Weeks [32.9% vs. Among both grDUps, the most common crises were 98 Deco-dents without known MHP had signi?cantly lower odds of recent release from any institution, but among 99 those who were recently released those without a known l?v?lHP were signi?cantly more likely to be released from a correctional facility [25.7% vs. or hospital (433% vs. 33.0%] than those with a known MHP. Among decedents with known MHP who were recently released from an institution 42.3% were IUI l02 released from facilities. [Decedents without known MHP, compared to those with were less likely to have a history of suicidal [04 ideation 95% 0:04-05) and prior suicide attempt 95% Both groups disclosed l05 suicide intent frequently, {22.4% vs. respectively] lilo Conclusions and Comments [6801?00 words] It)? From 1999?2015, 44 states saw significant suicide rate increases. Half of the states experienced increases of 30% or more. Rates increased significantly in 34 states among males and 43 states among females. Commant lzal?J]: I kept this paragraph con?ned to percentages but did add some additional aDR's in other places. Comment [15:19]: We were asked to include disclosure of suicide intent. something here like, This latter finding may indicate that the historical protective [Comment Izaf9]: Thinking to add effect of being female may be changinginto the causes of these trends is necessary 10 1' JL 1 JL ll? Researchers and practitioners regularly state that suicide is not caused by a single factor, however, the focus of II II suicide research and prevention practices, almost solely, focus on identify and treating MHP. The current study ?2 found that more than half of suicide decedents in did not have a known MHP. This group suffered more I I3 relationship problems and life stressors such as criminal-legal matters, evictionfloss of home, and recent or I I4 impending crises. This is particularly noteworthy in light of findings that suggest many suicides and attempts I IS occur with minimal deliberation I I6 Among people with MHP, two?thirds had a history of mental health andfor substance abuse treatment and over ll? half were in current treatment. This suggests that additional supports for this population are needed to keep HE them safe. This includes broader implementation of affordable and effective treatment modalities such as ?9 doctor-patient collaborative care models and cognitive-behavioral therapy. Additionally, greater access to [20 behavioral health providers, especially in underserved areas is needed, as is expansion of healthcare systems needed that integrate physical and behavioral health and that better support suicide prevention and patient Ill [22 safety, especially through care transitions 123 Study findings indicate that people with known MHP also experienced other life stressors such as joblfinancial, IE4 relationship, and/or physical health problems. These findings point to the need to both prevent the conditions Rev 315,18 v3.2 I25 associated with mental health problems in the first place and the need to support people with known MHP to decrease their risk of poor social, health, and economic outcomes These results, together, underscore the importance of comprehensive statewide suicide prevention activities I28 that address multiple factors associated with suicide. Prevention strategies may include: strengthening I29 economic supports le.g., housing stabilization policies, household financial support}; teaching coping and 130 problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belongingness and access to informational, tangible, 132 emotional, and social Support, and identifying and better Supporting people at risk. Other strategies include I33 creating protective environments leg, reducing access to lethal means among people at risk, creating I34 organizational and workplace policies to promote help-seeking, easing transitions into and out of Work for I35 people with MHP and other life challenges], supporting family and friends after a suicide, and assuring safe 136 reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning l3? and implementing such a comprehensive approach to suicide prevention I38 These findings have at least three limitations. In the state-level analysis, rankings for four states (MD, MA, RI, I39 might have been impacted by large proportions of injury deaths of undetermined intent lie. decreasing I40 suicide rates), or decreased percentages of such deaths over time lie. increasing suicide rates}. Second, is not yet nationally representative, the 2? states included in the current study of the U5. I42 hopulation. Third, abstractors of data are limited to information contained in investigative reports. anon-Ha I43 Therefore, the extent of informant knowledge can impact data completeness and accuracy. Studies including in? I44 depth interviews with next~ofal30% increases in 25 states. Mental health problems often contribute to suicide: however. 2015 data from the National Violent Death Reporting System [27 states} indicate 54% of suicide decedent's were not known to have problems. Other contributors included relationship. substance use. health. and jobi?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in CDC's Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025- [Comment How about saying ?lives were lost to suicide.. i i i i i . These need to be TABLE 1. Sele?ed demographic and descriptive characteristics of suicide: among aged 21!] years with and without known mental health problems National is . NIR Violent Heath 2? states 2015 5pc? IL (6.5, or new bummed Lu explgin. Total health problem' health problem I .. 1 039 P-iralue Comment Izni?Jl: ?aming. should this be, NA or Foomuted? Seems like NA would be a .m riale and easier, 6 BB . [1.91 2.3 2?2. 1 3420 3120 American I non-His Ever served in Incident a suicide suicides 54 0.3 Firean?n 9909 43. ueclass Other -ih 1 1 34.0 355 31.1 sanis 800 5 . 156 13. 2 Substance detected?? 1 . Positive 4 442 40.5 1. 0.01 00?0. Tested 0.Cocaine . E. . 0.01 0.9 0 SHDB 40.8 1?35 55.9 47'9 0.01 0.2 02:02 02 0240.3 intewai: o. Alaska. Arizona. Cotorado. Connecticut, Georgia. Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. Abbreviation: - New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. Decadent had been identified as having a current diagnosis of mental health problem in coronerlmedical examiner or law enforcement reports. 5 OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 'l Logistic regression was used to estimate adjusted OR with 95% Die after controlling for age. sex. race. and ethnicity. Known mental health problem was used as the reference group. Decedents Iivere aged 210 years. as per standard in the suicide prevention literature. Denominator is decedenls aged :18 years with reported military service status. it Denominator is deoedents who died by poisoning. including overdose. Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Adjusted on! {95% l'l't Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged 21D years with and without known mental health problems National Violent Death Reporting System. 2? probleml health problem no. no. slams 2015 Characteristic Mental health Current Problem 7' .2 1 1 . 1 15.2 424 4.5 Comment t; ditto above. 1 Comment Same thought. I i 5 i i Alcohoi substance mental healt'l'ii'suhstance abuse treatment treated for so nos 0 in month relatio robiem nonintimste Death of a loved one death fa stressars life stres sor Other robiem or loss home Recent release an or Crisis within or 2 wee Ra??i ?ner 5 L911 El note 501611.15 ll'llEl'll ideation 5141 .4 .B 94-8 42362 19.8 758 ,0 .0 .2 3 1'6?5 4-9.7 3?3 4.0 19 3182 2305 .8 .8 .5 .3 1 406 15.0 1 1 (0.01 14.1 <00 64 394 odds ratio; ?0.01 0.01 0. i? 0. :2 <0.01 0. 45.1 {0.01 . 1 <0.05 . 1 63 5.3 .01 5.1 2.099 152 3.0 New York. North Carolina. Ohio. Okiahorne. Oregon. Rhode Isiano?. South Carolina. Utah. Vermont. Virginia. and Wisconsin. 10.3 .D . . . costtreumatic stress disorder; SA suhstenoe abuse. ?0.01 1 .8 .01?001 .024] 1.3 1.2?1.3 1 . .4419 1.0 0.0?1.3?2. 0. 0.1-0.1 4 1. 1.4 1.3?1.5 1. 1 5?08 1.1 1.0?1.1 0.9 0.8-4.0 4 0.3 01?001 0.02?0.03 1.3 1.2-1.4 1. 1 1. 0.8 .51 1.1 34.3 1. 1 0.9 0.9 3-1Abbreviations: attention deficit disorderi'ettention de?cit hyperactivity disorder; 0R Alaska. Arizona. Gatorade. Connecticut, Georgia. Heweil. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. [zni?Sl]: These Four settings should he indented as they are :1 subset of the recent release from an institution. Recem release should remain unholded (saying this in case you were to hold it as with nthci' categories] as its part of ?Othcr LLife erCSsors? overarching category. lzal?il]: Same comment as above. these are subset of the recent crisis within the past or upcoming two and should he indented. Decedent had been identified as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health problem was the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. it Not significant. ?t Denominator is decedents aged :18 years. Tm Denominator is decedents aged 10?13 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonlv occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler. Scott R. Kegler. Kcming Yuan. Kristin lvi. Holland. Asha Z. Hey-Stephenson. Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015, a total of 54% of decedents in 27 states did not have a known mental health problem. Among decedents with infonnation on circumstances available. several circumstances were signi?cantly more likely among those without a known mental health problems than among those with mental health problems, including relationship problems/loss (45.1% versus life stressors (54.2% versus and crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. introduction in 2016, nearly 45,000 suicides (15.610 00,000 population [age-adjusted]) occurred in the United States among persons aged 210 years (I). From 1999 to 2015, suicide rates increased among both sexes, all racialr?ethnic groups, and all urbanization levels Suicide rates have also increased among all age groups vounaer than 75 years, with the highest percent increases among those aged 45?64 and those aged 10-14 Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing In addition, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased 42% from 2001 to 2016 Together, suicides and self?harm injuries cost the nation approximately $69 billion in direct medical and work less costs (I). The National Strategy for Suicide Prevention (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems, and preventing reattempts In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic problems, access to lethal means substances. firearms} among persons at risk, and poor coping and problem?solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help teach the national goal, established by the National Action Alliance for Suicide Prevention and the American Foundation for Suicide Prevention, reducing the annual suicide rates @2096 by 2025 (7). To assist states in achieving this goal. CDC analyzed state?speci?c trends in suicide rates and assessed the multiple contributing factors to suicide: this report presents options For multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 210 years only, as determining suicidal intent in younger children can be dif?cult Age?speci?c suicide counts were tabulated based on National 1vlital Statistics System coded death certificate records (International Classi?cation of Diseases, Tenth Revision. underlying-cause-of death Codes Y310, U03). Age-speci?c population estimates Were obtained from US. Census BureautNational Center for Health Statistics bridged~race population data releases. National and state?level suicide rate estimates were calculated for six consecutive 3?year aggregate periods spanning 999?2016 [1999?200l; 2002?2004; 2005?2007; 2008?2003; 2011?2013; and 2014- 201a]. Rate estimates were age-adjusted to the U.S. 2000 standard population and expressed per 100,000 persons per year. Age?adjusted suicide rate trends were modeled using the same 3?year data aggregates, employing weighted least?squares regression with inverse?variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics of persons aged 3:10 years who died by suicide, with and without known mental health problems. and the circumstances surrounding the suicides were compared in the 27 states* with complete data participating in CDC's National 1violent Death Reporting System in 20 5. defines mental health problems as disorders and listed in the Diagnostic and Statisticor' Marmot of Monte! Disorders. Fifth Edition (9), with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certificates, coronertmedical examiner reports (including toxicology], and law enforcement reports. A ranee ol?eircuntstunccs [relationship problems. lil'e stressors. and recenttimpendinu crises) has been identi?ed as potential risk factors for suicide. Circumstances captured bv are those identi?ed by next ot?kin as hating activelv to hnvg_ multiple circumstances. Decedents with and without known mental health problems were compared using Chi- square tests. Logistic regression analyses estimated adjusted odds ratios (aDRsl with 95% con?dence intervals controlling forage group. sex, and racetethnicity. Results The most recent overall suicide rates (representing 2014?201?) varied fourfold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period]. with absolute increases ranging from 0.8 per 100.000 {Delaware} to (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota], with increases >30% observed in 25 states (Supplementary Table; (Figure Modeled suicide rate trends indicated signi?cant increases in 44 states, among males {34 states) and; temales [43 states), as well as for the United States Comment (STU: This spacing looks odd. Are there options to ?x it? Nationally, the model-estimated AAPC for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated signi?cant AAPC increases for males and females {Supplementary Table; l. Suicide without known mental health problems {l 1,039) were compared with those w1th mental health problems {9.407) in 2'1 states. Whereas all decedents were predominately male {Table l) and non-Hispanic white those without known mental health problems. relative to those with mental health problems, were more likely to be male {33.6% versus odds ratio 95% CI and belong to a racialI'ethnic minority (OR range Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide? suicide Among adult dccedents aged 3:18 years 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever serv ed, or were currently serving. in the 1.1.3. military. Whereas firearms were the most common method of suicide overall and among decedents with and without mental health problems. without known mental health problems. relative to those with known mental health problems, were more likely to die by firearm {55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained sign itieant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results. without known mental health problems were less likely to test positive for any substance overall 0.3. 95% C1 including opioids 0.90. 95% C1 but were more likely to test positive for alcohol 1.2. 95%. C1 Information on circumstances surrounding suicide were available for all with mental health problems {11 9,40?) and approximately 85% of those without known mental health problems (n 9,357) in 2? states {Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7. 95% CI 0.7?0.8] than were persons with known mental health problems. Whereas two thirds of decedents with known mental health problems had a history of mental health or substance use treatment just over were in current treatment. Decedents without known mental health problems had a signi?cantly higher likelihood of any relationship problemf loss {45. than did those with known mental health problems speci?cally intimate partner problems {30.2% versus (17.5% versus and recently perpetrating interpersonal violence versus without known mental health problems were also more likely than were those with known mental health problems to have experienced any life stressors (54.2% versus 497%) such as criminalI? legal problems (10.2% versus or evictionlloss of home (4 3% versus and were more likely to have had a recent or impending [within the preceding or upcoming 2 weeks. respectively) crisis (a current or acute event thought to contribute to the suicide1{32.9% versus All of these differences remained signi?cant in the adjusted models. l'lLItlIll I I I. II. IIJ hl.? ll I. 111111 In IIcsIIt-cm II. .I+?Irmong all persons with recent crises. intimate partner problems were the most common types and did not differ by group. pl Il health Ins and ld' ms was tI-l [List-Ins health -II- and Ir'IHel? -lI1-h?pfob- 'rI?rl'l'l'lri- Decedents without known mental health problems had signi?cantly lower odds of recent release from any institution 0. 5 95% Cl? 4? ll. 51.; Among those recently released without known 1 . were signi?cantly more likely to have: mental health proble ms Ila . I . been released from a correctional facility {125.7% versus 8. hospital {43. 2% versus 33. or other; facility to g. alcohollsubstance treatment) than wete (- Comment IHTMU: So the ?rst number is the of 5.1 of with no known mental health issue who were recently release. Deleted S. I And, the second number is the of some of decedents with a known mental health issue who were recently released. correct? CORRECT r?ts I said on the phone, I had to read several times to ?gure who was who and where the percentages were coming from. ll-ltJl?LI' 1'5 NOW. OUT THE THE SINCE WE SAY AMONG "ll-105E Maybe something along the lines of below: without known mental health problems had signi?cantly lower odds of having been recently released from any institution 0.5, 95% CI Among those recently released from an institution. without a known mental health issue when compared with those with a known mental health problem Were signi?cantly more likely to have been released from a correctional facility {25.7% versus 3.2% 1, hospital (43.7% versus or other facility alcoholfsubstance treatmentthose recentlv released with a known mental health problem, respectivclv. Among decedents with known mental health problems who were recently released from an 46.7% were released from facilities. Decedents without known mental health problems were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% ofpersons without and with known mental health problems, respectively. Conclusions and Comments During 1999?2016, suicide rates increased significantly in 44 states, and 25 states experienced increases ?930%. Rates increased signi?cantly among males in 34 states, and females in 43 states. Illhis?odiag?is Midi Additional research into the specific causes of these trends is needed. Data from the 27 states participating in provide important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health problems, and prevention of reattempts. This study found that approximately half ofsuicide decedents in did not have a known mental health problem, indicating that additional focus on nonmental health factors further upstream could provide important information for a public health approach Those without a known mental health problem suffered more from relationship problems and other life stressors such as criminali?legal matters, evictionfloss of home, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as jobi?financial or physical health problems. These ?ndings point to the need to help persons both manage and prevent the conditions associated with mental health problems in the first place, and to support persons with known mental health problems to decrease their risk for poor outcomes Two thirds of this group with mental health problems had a history of treatment for any mental health or substance use or both. with approximately halfin treatment when they died. This finding suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities, such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. In addition, increased access to behavioral health providers in underscored areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (32). statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household ?nancial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicalfmental health problems) (I2). Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), strengthening access and deliver of care for people at risk. supporting family and friends after a suicide, and assuring the media follow safe reporting Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention The ?ndings in this report are subject to at least three limitations. First, in the state-level analysis, rankings for four states {Maryland, Massachusetts, Rhode Island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent {potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative: the 27 states included represent 49.6% of the population Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of-kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the. broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of POliCile-H, Programs, and Practices (I2) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard, Margaret Warner, Division of Vital Statistics, National Center for Health Statistics, CDC. Con?ict of Interest No con?icts ofinterest were reported. 1Division of'v'iolence Prevention. National Center for Injury Prevention and Control. CDC *?Division ofAnalysis. Rosearch, and Practice Integration. National Center for Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone. 270438?3942. References l. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. AZ, Crosby AE, Jack SPD, I-lailcyesus T, MJ. Suicide trends among and within urbanization levels by sex, racet?ethnicity, age group, and mechanism of death?United States, 2001?2015. MMWR Surveill Summ 2017;66{No, 83-18). 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. NCHS data bricf no 241. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS data brief no 293. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics: 2017. 5. Of?ce of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 Eaational strategy for suicide prevention: goals and objectives for action. Washington, DC: US Department: of Health and Human Services, Of?ce of the Surgeon General; 2012. rcport.pdf 6. Zalsman G. Hawton K. Wasserman. D. et Suicide prevention strategies revisited: Ill-year systematic review. Lancet Torguson K. O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; 2017. 8. Crepean-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2010;14:24u34. 9. American Association. Diagnostic and statistical manual of mental disorders 5). Washington. DC: American Association: 2.013. 10. Caine ED. Reed J. Hindinan J. Quinlan K. Comprehensive. integ'ated approaches to suicide prevention: practical guidance. Inj Prev 2017. Epub December 20. 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM. Holland KM. Bartholow BN. Crosby AE. Davis SP. Wilkins N. Preventing suicide: a technical package of pOllC?jiL'h. programs. and practice. Atlanta. GA: US Department of Health and Human Services. CDC: 2017. 13. Milner A. Sveticic J. De Lee D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. int Soc 14. Pouliot L. De Lee D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma, Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. Summary What is already known about this topic? In 2010. heady 45.000 lives Were lost to suicide in the United States. What is added by this report? During 1999?2016. suicide rates increased in nearly every state. including >30?i?u increases in 25 states. Mental health problems often contribute to suicide; however, 2015 data from the National Violent Death Reporting System states) indicate 54% of suicide decedents were not known to have such problems. Other contributors included relationship. substance use. health. and jobi?nanclai probiems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in CDC's Technical Package for Suicide Prevention. can help reach the national goal of reducing suicide rates 20% by 2025. znf'i All of these should be TABLE 1. Selected demographic and descriptive characteristics of suicide: among persons aged all] years with and without knawn mental health problems National KL: 0 I or n; as i111 Table 3 Violent [leath 2? states 2015 r' I new I Cumment HTM These needtnbe Total health problem' health problem on! i I (I 1 039 Pwame on? 9" more speci?c him. (animated to explain. a BB 1 3420 31:0 Amencanl . 11 non-His 4 2'10 Eversenredin . 1354 15.3 Incident la suicide suicides 54 0.3 Firearm 9909 43. class ca 1 1 34.0 800 5 Substance detented?? 1 . Positive 4 442 40.5 Positive 2.270 1.230 (20.1) i 1.041 0.01 i 0.0 (07?00)) 0.0 Benzodiazepines Tested 0.124(301} 4.2201440) 3.000 (35.3} 0.01 0.7 Positive 2.464 (30.3} 1.030 (38.8) 325 (21.2) 0.01 0.4 (0.4-0.5) 0.5 (0.5-0.0) Cocaine Tested rate (30.0) 3.060 (41.1) 4.112 [37.2) 0.01 0.010.040) 0.0 Positi'tte 400 (as) 210(501 203 (as) 0.05 1.2 (101.5) 1.2 Amphetamines Tested 7.015 {3722} 3.005 (30.3) 3.010 (35.5) 0.01 001103-00) 0.0 (0.0-0.0) Positive r30 are (10.2) 300 (0.2) ?rn 0.0 1.0 (0.0?1 .1) Ma_ri'yana Tested 5.500 {32.1} 3.1271332) 3,442 {31.2) 0.01 0.0 0.0 Positive 1.4?1 (22.4) T10 T61 (22.1) 1.0 (0.0-1.1) 0.9 (0.0-1.0) Antidepressants Tested 5.425 (26.5) 3.103 (33.0) 2.322 (21.0) 0.01 0.5 0.0 Positi've 2.214 (40.8) 1.735 (55.9) (20.6) 0.01 0.2 0.2 (0.2413) Abbreviation: CI con?dence interval: OR odds ratio. Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. Decedent had been identified as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. it DR reflects the risl-c among those without known mental health problems relative to those with known mental health problems. ii Logistic regression was used to estimate adjusted OR with 95% lills after controlling for age. sex. race. and ethnicity. Known mental health problem Was used as the reference group. Deoedents were aged 210 years. as per standard in the suicide prevention literature. 0' Denominator is decedents aged 213 years with reported military service status. it Denominator is decedents who died by poisoning, including overdose. 1" Denominator is decedents with any toxicology testi?ged. Denominator for each positive group is the number tested for the substance in that group. Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged :10 years with and without known mental health problems National Violent Death Reporting System. 2? states.* 2015 Known mental health No known mental problemt health problem P- Adjusted 0R1 Characteristic Total no. no. (fat) value [95% Cl) {35% Cl) Suicide with known circumstances 18.?64 (91.8) 9.40? (100) 9.35? (04.8) <0.01 Mental health Any Current Diagnosed Mental Health Problem? 0'52) Anxiety disorder 1.5?9 (16.3) Bipolar disorder 1.431 (15.2) Schizophrenia 509 - - - - PTSD 42:1- (4.5) ADDIADHD 226 (2.4) Comment Should these four settings be indented as they are a subset of Unknown Substance . the recent roloase frolu an Institution? Alcohol 58 1?.4 1 832 19.8 (0.01 [141.3 i Comment [2115']: Same question as l. above Currant marital substance abusa treatment 5 141 5 .0 . 1:001 0.01 01?0131 0.01 [Pl?0.01 ll var for mental 6 7'1? 35 323 .2 . <0.01 0.02 0242) [1.02 024} ll 1 3 125 4101 1.3 1.24. .3 1.24once ll in month 31 0.5 .4419 .nonintimate 201 1 .o 1 1.1 3-1.3 1 . ?1 Death of a loved one EN 118 0.9 3?0.9 'l Nonsuicida death 534 . 0.3 .1419 as 1. it or .8 Other life slressors life . . 5065 54.2 . ?l 1.1?1.3 1.1 1.1?1 . - - . I Other . . 1.0 .3-1 .1 1.0 .9home 31? 1. 1.141. . 1.241Reagent release from an institution?" 941 10.2 0.5 4?0. ll? 4?0. 5 "l on . 1 ~41211?! 1.3 1.0?1. 5} lotion 439 0.1 .1411 111 1 .1 1 1 . or Crisis within 2 vlissks1w 2 444 1.4 1.3-1.5 1' I . . 1 crisis .01 Le?anote 458 34 .8 1.1 1.0?1.1 Disclosed intent 4 05 23. .5 099 {0.Abbreviations: attention deficit disorderl'attention de?cit hyperactivity disorder: OR odds ratio; oosttrauma?o stress disorder; SA substance abuse. Alaska. Arizona. Colorado. Connsoliout, Georgia. Hawail. Kansas. Kentucky, Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jerssy. New Mexico. New York, North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. Decadent had been identi?ed as having a current diagnosis of mental health problem in coronerimedioal examiner or law enforcement reports. DR re?ects the risk among those without known mental health problems relative to those with known mental health problems. Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health problem was the reference group. Includes decedents with one or more diagnosed current mental health problems. which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. Not signi?cant. Denominator is decedents aged 218 years. Denominator is decedents aged 10?13 years. Denominator of institution subgroup is decedents with racenl release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonlyr occurring categories. FIGURE. Percentage change in annual suicide rate.? by state United States. from 1999?2001 to 2D14?2016cFigjmallaleigjmall:- The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 10D.DDO population. agevadjusted to the 21300 LLS. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler, Scott R. Kegler. Kenning Yuan. Kristin M. Holland. Asha Z. Hey-Stephenson. Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National 1'v?ital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states. with 25 states experiencing increases 80%. Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015, a total of 54% of decedents in 27 states did not have a known mental health problem. Among decedents with infonnation on circumstances available. several circumstances were signi?cantly more likely among those without known mental health problems than among those with mental health problems, including relationship problems/?loss {45.1% versus life stressors (54.2% versus and recentfimpending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. Introduction In 2016, nearly 45,000 suicides (15.6fl 00,000 population [age-adjusted]) occurred in the United States among persons aged 210 years From 1999 to 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide rates have also increased among all age groups younger than 75 years, with the highest percent increases among those aged 45-64 and those aged 10-14 (Lil). Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing In addition, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased 42% from 2001 to 2016 (I). Together, suicides and self?harm injuries cost the nation approximately $69 billion in direct medical and work less costs (1). The National Strategy for Suicide Prevention (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems, and preventing In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic problems, access to lethal means substances, ?rearms) among persons at risk, and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal, established by the National Action Alliance for Suicide Prevention and the American Foundation for Suicide Prevention, of reducing the annual suicide rate 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged E10 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (international Classi?cation of Diseases, Tenth Revision, underlying-cause?of death codes X60 X84. YETD. U03). Age?speci?c population estimates were obtained from US. Census BurcauINational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999?2016 (1999?2001; 2002?2004; 2005?2007; 2008?2010; 2011?2013; and 2014? 2016). Rate estimates were age-adjusted to the US). 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AA PCs). Characteristics of persons aged 210 Years who died by suicide, with and without known mental health problems, and the circumstances surrounding the suicides were compared in the 27 states* with complete data participating in CDC 's National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition with the exception ofalcoho] and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death ceni?catcs, coroner/ medical examiner reports (including toxicology), and law en foreemcnt reports. A range ofcircumstances (relationship problems, life stressors, and recent/?impending crises) has been identi?ed as potential risk factors for Suicide. Circumstances captured by are those identi?ed by next of kin as having actively contributed to a person?s suicide. Decedents could have experienced multiple circumstances. Decedents with and without known mental health problems were compared using Chi- square tests. Logistic regression analyses estimated adjusted odds ratios (aORs) with 95% confidence intervals (C15), controlling for age group, sex. and racefethnicity. Results The most recent overall suicide rates (representing 2014?2016) varied fourfold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period}, with absolute increases ranging from 0.8 per 100,000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases >30% observed in 25 states (Supplementary Table; (Figure Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females (43 states}, as well as for the United States overall (Supplementary Table; Nationally, the model-estimated for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated signi?cant AAPC increases for males and females (Supplementary Table; Suicide decedents without known mental health problems (11,039) were compared with those with known mental health problems (9,407) in 27 states. Whereas all decedents were predominately male (Table 1) and non-Hispanic white those without known mental health problems, relative to those with mental health problems, were more likely to be male (83.6% versus 68.8%; odds ratio 95% CI and belong to a racialfethnie minority (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide-suicide 2.9, 95% C1 Among adult decedents aged 3:18 years, 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever served, or were currently serving, in the US. military. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health problems, decedents without known mental health problems, relative to those with known mental health problems, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained signi?cant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (n 9,407) and approximately 85% of those without known mental health problems (n 9,357) in 27 states (Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% CI 0.7?0.8) than were persons with known mental health problems. Whereas two thirds of decedents with known mental health problems had a history of mental health or substance use treatment just over half were in current treatment. Deccdents without known mental health problems had a significantly higher likelihood of any relationship problemfloss than did those with known mental health problems speci?cally intimate partner problems (30.2% versus arguments/conflicts (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than were those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminalr?lcgal problems (10.7% versus or evictiontloss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Physical health problems and jobi?nancial problems were commonly contributing stressors among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.8%, respectively). Similarly, among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. without known mental health problems had signi?cantly lower odds of recent release from any institution 0.5, 95% CI Among those recently released, decedents without known mental health problems, were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcoholfsubstance treatment) (24.2% versus than those recently released with a known mental health problem, respectively. Among decedents with known mental health problems who were recently released from an institution, 46.7% were released from facilities. without known mental health problems were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health problems, respectively. Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states. and 25 states experienced increases Rates increased signi?cantly among males in 34 states, and females in 43 states. Additional research into the speci?c causes of these trends is needed. Data from the 27 states participating in provide important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health problems. and prevention of reattempts. This study found that approximately halfofsuicide in did not have a known mental health problem, indicating that additional focus on nonmental health factors further upstream could provide important information for a public health approach Those without a known mental health Problem suffered more from relationship problems and other life stressors such as eriminalflegal matters, eviction/loss of home, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as job/?nancial or physical health problems. These ?ndings point to the need to help persons both manage and prevent the conditions associated with mental health problems in the ?rst place, and to support persons with known mental health problems to decrease their risk for poor outcomes (l Two thirds of this group with mental health problems had a history of treatment for any mental health or substance use or both, with approximately halfin treatment when they died. This finding suggests the need for additional safety supports. including broader implementation of affordable and effective treatment modalities. such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. In addition, increased access to behavioral health providers in areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (12). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household ?nancial support); teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems. especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk te.g., military veterans, persons with physicalhnental health problems) Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), strengthening access and deliver of care for people at risk, supporting family and friends after a suicide, and assuring the media follow safe reporting recommendations {172). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention The ?ndings in this report are subject to at least three limitations. First, in the state-level analysis, rankings for four states (Maryland, Massachusetts, Rhode Island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent {potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 2? states included represent 49.6% of the population Finally. abstractors of data are limited to information contained in investigative repotts. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in? depth interviews with next-of-kin often identify greater attributions to mental disorders however. many methodological variations across studies exist It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown. and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policy. Programs, and Practices (12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgmems Robert Anderson. Holly Hedegaard. Margaret Warner. Division of Vital Statistics. National Center for Health Statistics. CDC. Conflict of Interest No conflicts ot?interest were reported. 'Division of Violence Prevention, National Center for Injury Prevention and Control. 3Division of Analysis. Research. and Practice Integration. National Center for Inj my Prevention and Control, CDC. Corresponding author: Deborah M. Stone. 270488?3942. References l. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta. GA: National Center for Injury Prevention and Control. Retrieved March 15. 2018. In. I'd Hey-Stephenson AZ, Crosby AE. Jack SPD, Haileyesus T. Kresnow-Sedacca MI. Suicide trends among and within urbanization levels by sex. race/ethnicity. age group. and mechanism of death?United States. 2001?2015. MMWR Surveill Stunm 2017;66tNo. SS-IS). 3. Curtin SC. Warner M. I-chegaard H. Increase in suicide in the United States. 1999?2014. NCHS data brief no 241. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2016. .pdf 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States. 2016. NCHS data brief no 293. Hyattsville, MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2017. 5. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action. Washington, DC: US Department of Health and Human Services. Of?ce ofthe Surgeon General; 2012. 6. Zalsman G. Hawton K. Wasserman D. et a1. Suicide prevention strategies revisited: 10-year systematic review. Lancet 2016;3z646v59. 7. Torguson K. O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; 2017'. 8. Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders 5). Washington, DC: American Association; 2013. 10. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Prev 2017. Epub December 20, 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package of policy, programs, and practice. Atlanta, GA: US Department of Health and Human Services, 2017. 13. Milner A, Sveticie J, De Lee D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 10. Alaska, Arizona, Colorado, Connecticut, Georgia, Ilawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan. Minnesota, New Hampshire, New Jersoy, Newr Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhoda island, South Carolina, Utah. Vennont, Virginia, and Wisconsin. Summary What is already known about this topic? In 2016, nearly 45,000 lives were lost to suicide in the United States. What is added by this report? During 1999?2016, suicide rates increased in nearly every state, including >30% increases in 25 states. Mental health problems often contribute to suicide; however, 2015 data from the National Violent Death Reporting System states) indicate 54% of suicide decedents were not known to have such problems. Other contributors included relationship, substance use. health, and jobl?nanciai problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in CDC's Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025. TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged :10 years with and without known mental health problems National Death Reporting System, 2? states) 2015 Known mental No known mental Total health printablem?r health problem Adjusted Characteristic (N 20,445) (n 9,40?) (n 11,039) F-value 0R5 (95% Cl) (95% Cl) Sex Male 15,?02 5,459 (58.8) 9,233 (33.5) 0.01 2.3 Female 4,744 (23.2) 2,935 (31.2) 1,305 (15.4) 0.01 0.4 Ase lyrsi? 10?24 2,304 1,211 (12.9) 1.593 (14.4) 0.01 1.1 25?44 5,455 (31.5) 3.035 (32.3) 3,420 (31.0) 0.05 0.9 45?54 3,820 (40.5) 3,398 (35.3) 0.01 0.5 255 3.458 1,340 (14.2) 2,125 (19.3) 0.01 1.4 (1.3-1.5) White, non-Hispanic 1?,102 (33.5) 3.155 (55.5) 5,93? (51.0) 0.01 0.5 Black, non-Hispanic 1,22?50) 411 51? GA) 0.01 American IndianIAlaska Native, non?Hispanic 3?8 (1.3) 112 (1.2) 255 (2.4) 0.01 2.0 Asian. non-Hispanic 575 (2.3) 235 (2.5) 341 (3.1) 0.05 1.2 (1.1-1.5) Hispanic 1.095 (5.4) 453 (4.9) 533 0.05 1.2 Other 55 (0.3) 21 (0.2) 45 (0.4) 0.05 1.5 Extended demographics Ever served in military,'11 3,429 1,354 (15.3) 2,0?5 (20.1) 0.01 1.4 1.1 (1.0?1 .1) Homeless 240 (1.2) 104 (1.1) 135 (1.3) ?10 1.1 1.2 (0.9-1.5) Incident Type Single suicide 20,053 (95.2) 9,315 (99.1) 10,?45 0.01 0.3 0.4 Homicide followed by suicide 319 (1.5) 54 255 (2.3) 0.01 3.5 2.9 Multiple suicides 54 (0.3) 25 (0.3) 39 (0.4) 1.3 1.5 Method Firearm 9.909 (43.5) 3.321 (40.5) 5.088 (55.3) 0.01 1.5 1.5 5,90? (25.9) 2,940 (31.3) 2,95? (25.9) 0.01 0.5 0.5 Poisoning 3,003 1,351 (19.8) 1,142 (10.4) 0.01 0.5 0.5 Substance class causing death?? Other over-the-counter) 1,021 (34.0) 555 (35.5) 355 (31.1) 0.01 0.5 0.9 Opioids 944 (31.4) 505 335 (29.4) 0.9 0.9 (0.3?1 .1) Antidepressants 300 (25.5) 544 (34.5) 155 0.01 0.3 0.3 Benzodiazepines 524 (20.5) 455 (25.1) 155 0.01 0.5 0.5 219 (7.3) 195 (10.5) 24 (2.1) 0.01 0.2 0.2 Other 1,595 750 (5.3) 515 0.05 0.9 0.9 Toxicology Results Any toxicology testing 5,555 (T05) 5,559 (50.3) 0.01 0.5 Positive for 21 substance?? 9,913 5,192 4,?21 0.01 0.5 Substance Alcohol Tested 10,950 (53.5) 5.409 5,541 (50.2) 0.01 0.5 Positive 4,442 (40.5) 2,115 (39.1) 2,32? (42.0) 0.01 1.1 1.2 Opioids Tested I 3,554 4.255 4,295 0.01 0.5 0.5 Positive I 2,2?9 1,238 1,041 0.01 0.8 0.9 Benzodiazepines Tested 3,124 4,226 (44.9) 3,898 (35.3) 0.01 Positive 2,464 (30.3} 1.639 (38.8) 325 (21.2) 0.01 0.4 0.5 Cocaine Tested (39.0) 3,866 (41.1) 4,112 0.01 0.9 0.9 Positive 499 215 (5.5) 283 (6.9) 0.05 1.2 1.2 Amphetamines Tested ?,615 3.696 (39.3) 3,919 (35.5) 0.01 0.9 (0.6-0.9) 0.9 Positive ?35 3?6 (10.2) 360 (9.2) 0.9 1.0 (03?1 .1) Marijuana Tested 6,559 (32.1} 3,12? (33.2) 3,442 (31.2) 0.01 0.9 0.9 Positive 1,4?1 (22.4) ?10 ?61 (22.1) 1.0 0.9 (0.6-1.0) Antidepressants Tested 5,425 (25.5) 3,103 (33.0) 2,322 (21.0) 0.01 0.5 0.6 Positive 2,214 (40.8} 1,?35 (55.9) 479 (20.6) 0.01 0.2 0.2 Abbreviation: CI con?dence interval; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decadent had been identified as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. it OR re?ects the risk among those without known mental health problems relative to those with known mental health problems. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race, and ethnicity. Known mental health problem was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. it Denominator is decedents aged :18 years with reported military service status. *4 Denominator is decedents who died by poisoning, including overdose. Til Denominator is decedents with any toxicology testing. Denominator for each positive group is the number tested for the substance in that group. Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged :10 years with and without known mental health problems National Violent Death Reporting System, 2? states,* 2015 Known mental health No known mental problem1 health problem P- Adjusted Characteristic Total no. no. (93) value 01239554 Cl) [95% Cl) Suicide with known circumstances 18.?64 (91.8) 9,40? (100) 9,35? (84.8) <0.01 Mental health Any Current Diagnosed Mental Health Problem? Anxiengisorder 1,5?9 (16.8) Bipolar disorder 1,431 (15.2) Schizophrenia 509 (5.4) PTSD 424 (4.5) ADDIADHD 225 Not speci?ed 780 (8.1) Current depressed mood 3,982 (42.1) 3,0?8 (32.9) <0.01 Substance problems Any Current substance problem 5,319 (28.3) 2,9?8 (31.8) 2.343 (25.0) <0.01 Alcohol problem 3.288 (1 1,882 (19.8) 1,405 (15.0) <0.01 Other substance problem 3,084 (18.4) 1,?88 (18.8) 1,318 (14.1) <0.01 Treatment Current mental healthisubstance abuse treatment 5,141 (54.0) 84 ?0.01 0.01 (0.01?0.01) 0.01 (0.01-0.01) Ever treated for mental healthisubstance problem (35.8) 8,323 394 (4.2) <0.01 0.02 (0.02?0.02) 0.02 (0.02-0.03) Relationship problemsiloss Any relationship problemiloss ?,948 (42.4) 3,?28 (39.8) 4,222 (45.1) <0.01 1.3 (1 1.3 (1 Intimate partner problem 5,098 2,2?0 (24.1) 2,828 (30.2) <0.01 1.4 (1 1.4 (1 Perpetrator of interpersonal violence in past month 414 (2.2) 131 (1.4) 283 (3.0) <0.01 2.2 2.0 (1.8-2.4) Victim of interpersonal violence in past month 84 (0.4) 53 (0.8) 31 (0.3) <0.05 0.8 (0.4-0.9) 0.8 Family relationship problem 1,8?1 (8.9) 8?3 (9.3) ?98 (8.5) 0.9 1.0 Other relationship problen?nonintimate) 403 (2.1) 202 (2.1) 201 (2.1) 1.0 1.1 Argument or con?ict (not speci?ed) 2,914 (15.5) 1,2?8 (13.8) 1,838 <0.01 1.3 1.4 Death of a loved one (any) 1,49? (8.0) 828 (8.8) 8?1 <0.01 0.8 0.9 (0.8-0.9) Nonsuicide death 1,181 (8.3) 84? (8.9) 534 <0.01 0.8 0.9 Suicide of family or friend 3?9 (2.0) 21? (2.3) 182 (1 40.01 0.8 Other life stressors Any life stressor 9,?43 (51.9) 4,8?5 5,088 (54.2) <0.01 1.2 (1 1.1 (1 Recent criminal legal problem 1,588 (8.5) 588 (8.2) 1.002 <0.01 1.8 (1.8-2.0) (1.5-1.9) Other legal problem ?48 (4.0) 3?8 (4.0) 3?0 (4.0) ?11 1.0 (0.8-1.1) 1.0 (0.9-1.2) Physical health problem 4,1?9 (22.3) 2,012 (21.4) 2,187 (23.2) 40.01 1.1 1.0 JobiFinancial problem?? 2941 (18.2) 1530 (18.8) 1411 (15.8) <0.05 0. 9 (0. 8?1. 0) 0.9 Eviction or loss of home ?22 (3.8) 31? (3.4) 405 (4.3) <0.01 1.3 1.4 School problemW 182 (19.9) ?0 92 (21.9) -11 1. 3 (0. 9?1. 8) 1.3 (0.9-1.9) Recent release from an 1,412 941 (10.2) 4?1 (5.1) <0.01 0 5 (0. 4?0. 5) 0.5 (0.4-0.5) JailiPrisoniDetention facility 203 (14.4) 82 121 40.01 3 8 (2 7?4. 9) 4.5 Hospital 51? (38.8) 311 (33.0) 208 <0.01 1. 8 (1. 3?2. 0) 1.3 hospitalfinstitution 489 (33.2) 439 30 (8.4) <0.01 0 1 (0.1-0.1) 0.1 Other (includes alcoholi'SA treatment facilities) 223 (15.8) 109 (11.8) 114 (24.2) <0.01 2. 4 (1. 843. 3) 2.5 (1.8-3.3) Recent or Impending Crisis Crisis within pastor upcoming 2 weekst?r?r 5,525 (29.4) 2,444 (28.0) 3,081 (32.9) ?0.01 1.4 (1 1.4 (1 Intimate partner problem crisis 1988 (35.8) 854 (34.9) 1114 (38.2) 1.1 1.1 Physical health problem crisis ?39 (13.4) 315 (12.9) 424 (13.8) 1.1 1.0 (0.8-1.2) Criminal legal problem crisis 821 (11.2) 203 (8.3) 418 (13.8) ?0.01 1.8 (1 Family relationship problem crisis 430 212 218 <0.05 0.8 0.9 Job problem crisis 354 (8.4) 191 183 (5.3) 40.01 Suicide eventJ'history Left a note 8,488 (34.5) 3,182 (33.8) 3.288 (35.1) 1.1 (1 1.2 (1 .1-1.2) Disclosed suicide intent 4,405 (23.5) 2,308 (24.5) 2,099 (22.4) 40.01 0.9 (0.8-1.0) 0.9 History of ideation 5,990 (31.9) 3,838 (40.8) 2,152 (23.0) <0.01 0.4 0.4 History of attempts 3,?32 (19.9) (29.4) 982 (10.3) <0.01 0.3 0.3 Abbreviations: ADDFADHD attention de?cit disorderi'attention deficit hyperactivity disorder; OR odds ratio; PTSD postlraumatic stress disorder; SA substance abuse. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health problems relative to those with known mental health problems. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health problem was the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. it Not significant. ?t Denominator is decedents aged :18 years. Tm Denominator is decedents aged 10?13 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonlv occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Comment Ll. Please replace ?mental health problems" with ?mental health conditions? throughout. By de?nition in methods. these are medical conditions. Calling them ?problems" trivializcs them and adds further stigma because a person should be able to cope with their problems in our culture. Most are conditions that can be medically treated. Also, is a the seienti the fact sheet, and all but the should be written in scienti?" Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler. Scott R. Kegler. Yuan. Kristin M. Holland. Asha Z. It-?ey-Stephenson. Pth: Alex E. CrosbyAbstract Introduction: Suicide rates in the United States have risen nearly 30% since WW. and mental health Ii bonditions?are just one factor contributing to suicide. Examining state-level trends in suicide and the It I Happy to discuss with a 't I 1 i multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. 1 Methods: Trends in agesadjustcd suicide rates among persons aged 3:10 years. by state and sex, across six consecutive 3-year periods (1999?2016}, were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia Data from the National rViolent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions- Results: During 1999?20l6, suicide rates increased significantly in 44 states, with 25 states experiencing increases Rates increased significantly among males and females in 34 and 43 states. respectively. [n 2015, atetalre??t?i?it: of decedents in 27 states did not have a known mental health condition. Among dccedents with available infonnation, several circumstances were signi?cantly more likely among those without a known mental health condition than among those with mental health condition, including relationship life stressors {253% versus __a_t;tgl_ crises {32.9% versus Put these circumstances were common across groups I I Conclusions: Suicide rates increased signi?cantly across most states during Egg?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. I .5 1" Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs. identify and support persons at risk. prevent reattempts. and help friends and family in the aftermath of a suicide. i't Introduction In 2Ul6, nearly 45.0tltl suicides population [age-adjusted?: ocCurred in the United States among persons aged Ell] years From 1999 to 2015, suicide rates increased among both sexes. all raciah?ethnic groups, and all urbanization levels Suicide rates also increased among all age erouns Comment From Michael Iadcrnarco: L2. Stigmatizing language. I don?t want to rock the boat so late int eh Ldiagnoses. Comment Mental health Comment KC L3. Find contusin_. "1 process. Are these mental health "problems?" I want to be sure we are not using stigmatizing language. How about "conditions?? This occurs throughout the manuscript. In Methods, it seems the data source will support the use of the word conditions and condition different than relationship problem even if both can be devastating. Comment infill: The original ordering here is preferred. The circumstances noted arejust three of many. We?d like to highlight the di??crences between people with and without mental health conditions while at the same time noting that these factors are actually very important to both younger than 75 years. with the highest percent increases among those aged ill-141769!" increase I from LE to 2.1 per IUUDOU in 1999 and 2016. respectivelvll and those aged 45764 [45% increase [from 13.2 to I?ll per 100300 In l999 and 2016. respectivelylj. Suicide is the 10th leading cause of death and is one ofjust three leading causes that are increasing (L4). In addition. rates of emergency department visits for nonfatal self-harm. a main risk factor for suicide, increased 42% from 2001 to 2016 Togethen suicides and self- harm injuries cost the nation approximately $69 billion in direct medical and work loss costs (I). The National Strategy for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels (in? individual1 t?amilyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide Ls rarely caused by any single factor, but rather, is determined by multiple factors. Despite NSSP guidance, suicide prevention largely focuses on n" identifying suicidal persons. providing treatment for mental health {conditions}. and preventing reattempts I. I l' I ?59325- Comment Please go through rest ofre on and chan e. A Comment izaf9l: Done! In addition to mental health conditions and prior suicide attempts, other contributing circumstances include social and economic problems, access to lethal means substances, ?rearms] among persons at risk, and poor coping and problem?solving skills Expanded awareness of these additional CirCumstanccs contributing to suicide risk and action to address them can help reach the national goal established by the American Foundation for Suicide Prevention and the National Action Alliance of Suicide Prevent ofrcducing suicide rates by 20% by 2025 To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 3:10 years only, as determining suicidal intent in younger children can be difficult Age?speci?c suicide counts were tabulated based on National 1t?ital Statistics System coded death certi?cate records (international Classi?cation of Driven-res, Tenth Revision, underlying-cause-ofdeath codes Xotl?th?l, U03). Age-speci?c population estimates Were obtained from LLS. Census BureauiNational Center for Health Statistics bridged~race population data releases. National and state?level suicide rate estimates were calculated for six consecutive 3?year aggregate periods spanning 999?201? [1999?200l; 2002?2004; 2005?2007; 2008?20l0; 2011?2013; and 2014? mm]. Rate estimates were age-adjusted to the U.S. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3?year data aggregates. employing weighted least-squares regression with inverse?variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Comment IKCU: L3. Because used only est 5 ell throu hour. 1-1 more times. so 1 Comment leat?ll: Removed acronym in all three instances where it appeared. Comment IKCU: L3. To make it easier for reader, please de?ne abbreviations for states in this table. Some people might not its have all 50 abbreviations in their heads. ll ll ll Comment Will add de?ne i abbreviations in the sopplenientatjt,r table._, Characteristics of persons aged 1:10 years who died by suicide, with and without known mental health conditions. and the circumstances surrounding the suicides were compared in the 2? states?? with complete data participating in CDC's National Violent Death Reporting System in 20 5. WDRS defines mental health conditions as disorders and listed in the Diagnostic and Statistics?! Manual of Mental Disorders. Edition with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certificates, examiner reports {including toxicology}, and law enforcement reports. Decedents with and without known mental health conditions were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios (aORsi with 95% confidence intervals lifts). controlling for age group, sex, and racei'ethni city. Results The most recent overall suicide rates {re resenting 20l4?201?) varied fourfold, from 6.9 (DC) to 29.2 [Eupplementaryl Table; 1'85}, (Montana) per 100,000 persons per year Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period], with absolute increases ranging from 0.8 per 100,000 {Delaware} to {Wyoming}- Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota]. with increases >30??o observed 25 states (Supplementary Table; overall (Figure?l). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males {34 states] and (Supplementary Table; as for the United States Well females [43 states), as Nationally, the model?estimated AAllGai'eruec annual percentage change for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated significant average annual percentage change increases For males and females {Supplementary Table; Suicide decedents without known mental health conditions (11,039) were compared with those with known mental health conditions (9,407) in 27 states. Whereas all decedents were predominately male (Table l) and non-Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 95% CI and belong to a racialfethnic minority (OR Suicide decedents without known mental health conditions also had signi?cantly higher odds of perpetrating homicide-suicide 2.9, 95% CI Among adult decedents aged 218 years, 20.1% of those without known mental health conditions and 15.3% ofthose with mental health conditions had ever served in the US. military or were serving at the time ofdeath. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health conditions, decedents without known mental health conditions, relative to those with known mental health conditions, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained significant in the adjusted models. Toxicology testing was less likely to be performed for decedents without known mental health conditions. Among those with toxicology results, decedents without known mental health conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, C1 Information on circumstances surrounding suicide were available for all decedents with mental health conditions (n 9,407) and approximately 85% of those without known mental health conditions (n 9,357) in 27 states (Table 2). Persons without known mental health conditions were less likely to have any substance use disorders 0.7, 95% CI 0.7?0.8) than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment just over half were in treatment at the time of death. Decedents without known mental health conditions had a significantly higher likelihood of any relationship than did those with known mental health conditions speci?cally intimate partner problems (30.2% versus argumentsi?con?icts (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (54.2% versus 49.7%) such as criminalflegal problems (10.7% versus or eviction/loss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained significant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobr??nancial problems were commonly experienced among both persons without mental health conditions (23.2% and 15.6%, respectively) and those with mental health conditions (21.4% and 16.8%, respectively). Decedents without known mental health conditions had significantly lower odds of recent release from any institution 0.5, 95% CI but those who were recently released were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcohoUsubstance treatment) 2.5 95% C1 than were those with a known mental health condition. Among decedents with known mental health conditions who were recently released from an institution 46.7% were released from facilities. Decedents without known mental health conditions were significantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health conditions {40.3% and 29.4%. respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health conditions. respectively. Conclusions and Comments During l999--20 6. suicide rates increased significantly in 44 states. and 25 states experienced increases 30%. Rates increased significantly among males in 34 states. and females in 43 states. This ?nding is consistent with prior research that indicated a decreasing gender gap in male-female suicide rates duringr l999?2014 Additional research into the speci?c causes of these trends is needed. Data from the 2? states participating in provide important insight into circumstances surrounding suicide and can i I help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor (5): however. suicide prevcnliort I is often oriented toward identi?cation of suicidal persons, treatment of mental health conditions. and prevention of This study found that approximately half of suicide in did not have a knovvn mental health condition. indicating that additional focus on nomnental health ti factors further upstream could provide important information for a public health approach (ill). Those without a known mental health condition suffered more from relationship problems and other life stressors - h?u? such as matters. evictionlloss of home. and recent or impending crises. .9: Similarly. persons with mental health Conditions often experienced relationship problems and other life i if stressors such as jobr??nancial or physical health problems. (These ?ndings point to the need to help persons ii both manage and prevent the conditions associated with mental health conditions in the ?rst place. mild to i l; 3 support persons with known mental health conditions to decrease their risk for poor outcomes 1). up} thirds ot?suicide decedents with mental health conditions had a history of treatment for any mental health .i or substance use or both. with approximately half in treatment when they died. This ?nding suggests the i need for additional safety supports. including broader implementation of affordable and effective treatment 3 modalities. such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. In 5 addition. increased access to behavioral health providers in underserved areas is needed. as is expansion of health care systems that integrate physical and behavioral health. with a priority on suicide prevention and i i i i i i patient safety. especially through care transitions I Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies. household ?nancial support): teaching coping and problem?solving skills to manage everyday stressors and prevent future relationship problems. especially early in life; promoting social connectedness to increase a sense of belonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk military veterans. persons with physicalr?mental health conditions) Other strategies include creating protective environments leg. I reducing access to lethal means among persons at l?isld of suicide. creating organizational and workplace policies to promote help-seeking. easing transitions into and out of work for persons with mental health conditions and other life challenges). supporting fantily and friends after a suicide. and assuring the media follow safe reporting guidelines Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention Hi). The ?ndings in this report are subject to at least three limitations. First. in the staterlevel analysis. rankings for four states (Maryland. Massachusetts. Rhode island. and Utah} might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second. is not yet nationally representative; the 2? states included represent 49.6% of the Comment L2. Has this been reviewed by a Not sure what \is meant by "prevent the conditions?- i 1' Comment [lam]: This hasn't been .I reviewed by a but we know 5' that certain circumstances such as adverse 5 childhood experiences are associated with ?lture mental health conditions so we want to also prevent those conditions from . occurring in the ?rst place. f??i kL3. MMWR. of self-inflicted i it {Comment loaf'il: Prefer to say suicide Comment Michael lademarco: here vs self-in?icted iniury. Finally. population absn'actors of data are limited to information contained in investigative reports. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in? depth interviews with next-of-kin often identify greater attributions to mental disorders however. many methodological variations across studies exist It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown. and hence underreported by key informants. Nonetheless. the high prevalence of diverse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies. Programs. and Practices to better understand suicide in their eemmanitiespopulations. prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgmta-nts Robert Anderson. Holly Hedegaard. Margaret Warner. Division of Vital Statistics. National Center for Health Statistics. CDC. Conflict of Interest No con?icts ot?interest were reported. 'Division of Violence Prevention. National Center for Injury Prevention and Control. 3Division of Analysis. Research. and Practice Integration. National Center for Inj my Prevention and Control. CDC. Corresponding author: Deborah M. Stone. dstone3@cdc.gov. 170?488?3942. References l. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta. GA: National Center for Injury Prevention and Control. Retrieved March 15. 201 S. In. I'd Hey-Stephenson AZ. Crosby AE. Jack SPD. Haileyesus T. Kresnow-Sedacca MJ. Suicide trends among and within urbanization levels by sex. race/ethnicity. age group. and mechanism of death?United States. 2001?2015. MMWR Surveill Stunm 2017;66tNo. SS-IS). 3. SC. Warner M. I-Iedegaard H. Increase in suicide in the United States. 1999?2014. NCHS data brief no 241. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2016. .pdf 4. Kochanek K. Murphy S. Xu J. Arias E. Mortality in the United States. 2016. NCHS data brief no 293. Hyattsville. MD: US Depaitment of Health and Human Services. CDC. National Center for Health Statistics; 20l7. 5. Of?ce of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 national strategy for suicide prevention: goals and objectives for action. Washington. DC: US Department of Health and Human Services. Of?ce of the Surgeon General; 2012. 6. Zalsman G. Hawton K. Wasserman D. et a1. Suicide prevention strategies revisited: 10-year systematic review. Lancet 2016;3z646v59. 7. Torguson K. O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; 2011'. 3. Crepeau-Hobson .F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association Diagnostic and statistical manual of mental disorders 5). Washington, DC: American Association; 2013. 10. Caine ED, Reed Hindman J. Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Prev 2017. Epub December 20. 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva. Switzerland: World Health Organization: 2012. 12. Stone DM. Holland KM. Bartholow BN. Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package of policies. programs, and practice. Atlanta. GA: US Department of Health and Human Services. 2017. 13. Milner A. Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. lnt.l Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona. Colorado, Connecticut, Georgia, Hawaii. Kansas. Kentucky. Maine, Maryland, Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode island. South Carolina, Utah. Vermont. Virginia. and Wisconsin. Summary What is already known about this topic? In 2016. nearly 45.090 deaths?Islese?eeusedgeogle died by suicide in the United States. Mental health conditions can contribute to suicide. What is added by this report? During 1999?2919. suicide rates increased in nearly every state. including increases in 25 states._4v1ental health data from the-National?Vieientaealh Reporting System (-27 states-i indicate 54% of suicide decedents were not known to have mental health conditions. Other contributors included relationship. substance use. health. and jobl?nancial problems. among others. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in CDC's Technical Package for Suicide Prevention. can help reach the national goal of reducing the annual suicide rate by 2025. TABLE 1. Selocted demographic and descriptive characteristics of suicide: among parsons aged all] years with and without known mental health conditions National Violent [loath 2? states 2015 American I non-His Ever served in Incident a suicide suicides Fimarrn no class Other -ih sanls 2 Substance detected?? Positive 1 nown Total health can clilinnIr 3 health condition I 1 039 6 SB 1 3420 31:0 54 0.3 9909 43. ID 355 31.1 156 13. 4442 40:5 [1131 2.3 2?2 [will]: This footnote also 3' applies in the other dashes in this column. a Not sure if you want the dashes to remain or not 5n lo? than] in. Positive 2.2?3 (23.3? 1.233 (23.111 1.041 0.01 03107?03)! 03103?10) Benzodiazepines Tested 3.124 4.2231443) 3.333 135.3} 0.01 0.7 Positive 2.434 (30.3} 1.333 (33.3) 325 (21.2) 0.01 0.4 0.5 (0.5-0.3) Cocaine Tested 7.333 (33.0} 3.333 1.41.1) 4.112 [33.2) 0.01 0.3103303) 0.3 Positive 43313.3) 213 (3.3) 233 (3.31 0.05 1.2 1.2 Amphetamines Tested 7.313 {3712} 3.333 (33.3) 3.313 135.3) 0.01 031100-03) 03103-03) Positive 333.133} 373 (102) 330 (3.2) ?111 0.3 1.0 (0.3?1 1) Ma_riiuana Tested 3.5331321} 3.1271332) 3.442 {31.2) 0.01 0.3 0.3 Positive 1.4?1 {22.41 T10 T61 {22.1) -W 1.0 (0.9-1.1) GENE-1.01 Antidepressants Tested 5.425 {26.5) 3.103 (33.0) 2.3221210) 0.01 0.5 0.0 Positi've 2.2140103} 1.735 (55.91 471120.61 0.01 0.2 [32?02) 0.2 Abbreviation: CI con?dence interval: OR odds ratio. Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South lCarolina. Utah. Vermont. Virginia. and Wisconsin. Decadent had been identified as having a current diagnosis of mental health condition in coronerimedical examiner or law enforcement reports. *5 OR reflects the rial-c among those without known mental health condition relative to those with known mental health condition. 1 Logistic regression was used to estimate adjusted OR with 95% l[Ills after controlling for age. sex. race. and ethnicity. Therefore. adjusted OR for age groups. sex and racei?ethnlolty groups are not presented. Known mental health condition was used as the reference group. Decedents were aged 310 years. as per standard in the suicide prevention literature. Denominator is decedents aged 213 years with reported military service status. Denominator is deoedents who died by poisoning. including overdose. '11 Denominator is decedents with any toxioology tested. Denominator for each positive group is the number tested for the substance in that group. TH Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? state-s," 2015 Known mental health No known mental health condition1 condition P- Adjusted DR'll (95% Characteristic Total no. no. value OR5 [95% Cl} 131) Suicide with known circumstances 18.704 {91.8) 9.40? (100] 9.35? {34.3} <0.01 Mental health Any Current Diagnosed Mental Health Condition? TING ?5.21 N?w Anxiety disorder 1.5?9 {15.81 Bipolar disorder 1.431 (15.2] Schizophrenla 509 (5.4) PTSD 424 226 Unknown 760 {3.1 Current depressed mood 3,962 {42.1} 3.076 {32.9} 40.01 0.7 0.7 Substance problems Any Current substance problem 5.319 {23.3} 2.976 {31.6} 2,343 {25.0} {0.01 0.7 0.7 Alcohol problem 3.266 {17.4} 1.362 {19.6} 1,406 {15.0} (0.01 0.7 0.7 Other substance problem 3,064 {16.4} 1.763 {16.3} 1,316 {14.1} <0.01 0.7 0.7 Treatment Current mental abuse treatment 5,141 {27.4} 5.077 {54.0} 64 <0.01 0.01 {0.01?0.01} 0.01 {0.01-0.01} Ever treated for mental healthisubstance problem 6,717 {35.3} 6,323 {67.2} 394 ??0.01 0.02 {0.02?0.02} 0.02 {0.02?0.03} Relationship problemsiloss Any relationship problemfioss 7.946 {42.4} 3.726 {39.6} 4.222 {45.1} <0.01 153312?13} 1.3 {1 Intimate partner problem 5,096 {27.2} 2.270 {24.1} 2.626 {30.2} r?0.01 1.4 1.4 Perpetralor of interpersonal violence in past month 414 131 233 <0.01 2.2 2.0 Victim of interpersonal violence in past month 34 53 31 {0.05 0.6 0.3 Family relationship problem 1.67116.9)_ 673 79343.5} 0.9 Other relationship problem {nonintimate} 403 202 201 1.0 1.1 Argument or con?ict {not speci?ed} 2,914 (15.5} 1,276 {13.6} 1,636 {17.5} (0.01 1.3 1.4 {1 -3-1 Death ofa loved one {any} 1.497 326 671 <0.01 0.6 0.9 Nonsuicide death 1.131 647 534 <0.01 0.3 0.9 Suicide of famin or friend 379 217 162 4:10.01 0.7 0.3 Other life stress-ore Any life stressor 9.743 {51.9} 4,675 {49.7} 5.066 {54.2} {0.01 1 2 1.1 {1 Recent criminal legal problem 1.566 566 1.002 {10.7} <0.01 1 6 1.7 Other legal problem 746 373 370 ?11 1.0 1.0 Physical health problem 4,179 {22.3} 2.012 {21.4} 2,167 {23.2} <0.01 1.1 {1 1.0 {1 JobiFinancial problem?? 2941 {16.2} 1530 {16.6} 1411 {15.6} (0.05 0.9 0.9 Eviction or loss of home 722 317 405 s0.01 1.3 1.4 School problem?P'IT 162 {19.9} 70 {17.3} 92 {21.9} ?Tt 1.3 1-3 Recent release from an institutionm 1,412 941{10.2} 471 <0.01 0.5 {04-05} 0.5 JailiPrisoniUetention facility 203 {14.4} 32 121 {25.7} <0.01 3.6 45332?64} Hospital 517 {36.6} 311 {33.0} 206 {43.7} <0.01 1.6 1.3 {1 hospitaliinstitution 469 {33.2} 439 {46.7} 30 <0.01 0.1 0.1 Other (Includes alooholiSA treatment facilities) 223 {15.3} 109 {11.6} 114 {24.2) <0.01 2.4 2.5 Recent or Impending Crisis Crisis within pastor upcoming 2 5,525 {29.4} 2,444 {26.0} 3,061 {32.9} <0.01 1.4 1.4 Intimate partner problem crisis 1966 (35.6} 654 {34.9} 1114 {36.2} 1.1 1.1 Physical health problem crisis 739 (13.4} 315 {12.9} 424 {13.3} 1.1 1.0 Criminal legal problem crisis 621 {11.2} 203 416 {13.6} <0.01 1.7 1.6 FamilyI relationship problem crisis 430 212{?7} 213 410.05 0.3 0.9 Job problem crisis 354 191 163 40.01 0.7 0.7 Suicide eiirentfhistoryr Left a note 6.466 {34.5} 3.162 {33.6} 3.266 {35.1} ?71 1.1 1.2 {1 Disclosed suicide intent 4,405 {23.5} 2.306 {24.5} 2.099 {22.4} <0.01 0.9 0.9 History of ideation 5,990 {31.9} 3,636 {40.6} 2.1 52133.0} (0.01 0.4 History of attempts 3,732 {19.9} 2.770 {29.4} 962 {10.3} <0.01 0.3 0.3 Abbreviations: attention de?cit disorderiatlention deficit hyperactivity disorder; OR odds ratio: PTSD postlraumatic stress disorder: SA substance abLise. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South lCarolina, Utah, Vermont, Virginia, and Wisconsin. Decadent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. *3 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OR with 95% l[Sis after controlling for age, sex, race, and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. YT Not significant. Denominator is decedents aged 3:18 years. Denominator is decedents aged 10?18 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonly occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler, Scott R. Kegler. Kcming Yuan. Kristin M. Holland. Asha Z. Hey-Stephenson. Alex E. Crosby, MIDI1 Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can infomi comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persoms aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National 1'v?ital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 20l5, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states. with 25 states experiencing increases P3093. Rates increased significantly among males and females in 34 and 43 states, respectively. in 2015. 54% of decedents in 27 states did not have a known mental health condition. Among decedents with. available information, several circumstances were signi?cantly more likely among those without a known mental health condition than among those with mental health condition, including relationship problemsfloss (45.1% versus life stressors {54.2% versus and recentx?impending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. Introduction In 2016, nearly 45,000 suicides (15.6? 00,000 population [age-adjustedD occurred in the United States among persons aged 210 years From 1999 to 2015, suicide rates increased among both sexes, all racial/ethnic groups, and all urbanization levels Suicide is the 10th leading cause ofdeath and is one of just three leading causes that are increasing (L4). In addition, rates of emergency department visits for nonfatal self-harm. a main risk factor for suicide. increased 42% from 2001 to 2016 Together, suicides and self?harm injuries cost the nation approximately $69 billion in direct medical and work loss costs The National Strategy for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familys?relationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite NSSP guidance. suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health conditions, and preventing reattempts (6). In addition to mental health conditions and prior suicide attempts, other contributing circumstances include social and economic problems, access to lethal means substances. ?rearms) among persons at risk. and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal established by the American Foundation for Suicide Prevention and the National Action Alliance of Suicide Prevent ofreducing suicide rates by 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-specific trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged I310 years only, as determining suicidal intent in younger children can be dif?cult Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (International Cfossi?cotion of Diseases, Tenth Revision. nnderlying-cause-of death codes Y87.0, U03). Age-speci?c population estimates were obtained from US. Census Bureaquational Center for Health Statistics bridged?race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999?2016 (1999?2001; 2002?2004; 2005?2007; 2008?2010; 2011?2013; and 2014? 2016). Rate estimates were age-adjusted to the U.S. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes. Characteristics of persons aged :10 years who died by suicide, with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 27 states* with complete data participating in National Violent Death Reporting System in 2015. de?nes mental health conditions as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance use disorders. which are captured separately in aggregates data from three primary data sources: death certificates, coronerimedical examiner reports {including toxicology), and law enforcement reports. Decedents with and without known mental health conditions were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios (aORs) with 95% con?dence intervals (Cls). controlling for age group, sex, and race/ethnicity. Results The most recent overall suicide rates (representing 2014?2016) varied fourfold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from 0.8 per 100,000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases 380% observed in 25 states (Supplementary Table: (Figure). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females {43 states), as well as for the United States overall (Supplementary Table; Nationally, the model-estimated average annual percentage change for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated signi?cant average annual percentage change increases for males and females (Supplementary Table; Suicide decedents without known mental health conditions (11,039) were compared with those with known mental health conditions (9,402) in 27 states. Whereas all decedents were predominately male {Table l) and non-Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 95% and belong to a minority (OR range= Suicide decedents without known mental health conditions also had signi?cantly higher odds of perpetrating homicide-suicide 2.9, 95% C1 Among adult decedents aged 318 years, 20.1% of those without known mental health conditions and 15.3% of those with mental health conditions had ever served in the U.S. military or were serving at the time of death. Whereas ?rearms were the most common method of suicide overall and among with and without mental health conditions, dcecdents without known mental health conditions, relative to those with known mental health conditions, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained signi?cant in the adjusted models. Toxicology testing was less likely to be performed for decedents without known mental health conditions. Among those with toxicology results, without known mental health conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, CI Information on circumstances surrounding suicide were available for all decedents with mental health conditions (n 9,407) and approximately 85% of those without known mental health conditions (11 9,357) in 27 states (Table 2). Persons without known mental health conditions were less likely to have any substance use disorders 0.7, 95% CI 0.7?0.8) than were persons with known mental health conditions. Whereas two thirds of dccedents with known mental health conditions had a history of mental health or substance use treatment just over half were in treatment at the time of death. Decedents without known mental health conditions had a signi?cantly higher likelihood of any relationship problem/loss than did those with known mental health conditions speci?cally intimate partner problems (30.2% versus arguments/conflicts (17.5% versus and recently pcrpetrating interpersonal violence versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (54.2% versus 49.7%) such as criminalx?lcgal problems (10.7% versus or eviction/loss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and job/financial problems were commonly experienced among both persons without mental health conditions (23.2% and 15.6%, respectively) and those with mental health conditions (21.4% and 16.8%, respectively). Decedents without known mental health conditions had signi?cantly lower odds of recent release from any institution (210R 0.5, 95% C1 but those who were recently released were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcohol/substance treatment) 2.5 95% C1 than were those with a known mental health condition. Among decedents with known mental health conditions who were recently released from an institution 46.7% were released from facilities. Decedents without known mental health conditions were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health conditions, respectively. Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increases 360%. Rates increased signi?cantly among males in 34 states, and females in 43 states. This ?nding is consistent with prior research that indicated a decreasing gender gap in male-female suicide rates during 1999?2014 Additional research into the specific causes of these trends is needed. Data from the 27 states participating in provide important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health conditions, and prevention of rcattempts. This study found that approximately half of suicide decedents in VDRS did not have a known mental health condition, indicating that additional focus on nonmental health factors further upstream could provide important information for a public health approach Those without a known mental health condition suffered more from relationship problems and other life stressors such as criminal/legal matters, evictionr?loss ofhome, and recent or impending crises. Similarly, persons with mental health conditions often experienced relationship problems and other life stressors such as jobffinancial or physical health problems. These ?ndings point to the need to help persons both manage and prevent the conditions associated with mental health conditions in the first place, and to support persons with known mental health conditions to decrease their risk for poor outcomes (Hi. Two thirds of suicide decedents with mental health conditions had a history of treatment for any mental health er substance use or both, with approximately half in treatment when they died. This ?nding suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities, such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. in addition, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions U2). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household financial support): teaching coping and problem-solving skills to manage everyday stressors and prevent ?xture relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicalimental health conditions) Other strategies include creating protective environments reducing access to lethal means among persons at risk of suicide, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health conditions and other life challenges), supporting family and friends after a suicide, and assuring the media follow safe reporting guidelines (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention (1 U). The ?ndings in this report are subject to at least three limitations. First, in the state-level analysis, rankings for four states (Maryland, Massachusetts, Rhode Island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent {potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of-kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown, and hence underreported by key infonnants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies, Programs, and Practices to better understand suicide in their populations, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly .Hedegaard. Margaret Warner, Division of Vital Statistics, National Center for Health Statistics. CDC. Con?ict of Interest No conflicts of interest were reported. 'Division ol'Violence Prevention. National Center for Injury Prevention and Control, CDC: 1Division ofAnalysis. Research. and Practice Integration, National Center for I njury Prevention and Control. CDC . Corresponding author: Deborah M. Stone, dstonelit?gimdcgov. 770?438?3942. References l. CDC. Webwbascd Injury Statistics Query and Reporting System Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2013. In. 2. Ivey?Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow-Sedacca MJ. Suicide trends among and within urbanization levels by sex, racet'cthnicity. age group, and mechanism of death?United States, 2001?2015. MMWR Surveill Sumrn 201?;66tNo, SS-IS). 3. Curtin SC. Warner M. Hedegaard H. Increase .in suicide in the United States, 1999? 2014. NCHS data brief no 241. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. .pdf 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS data briefno 293. Hyattsville. MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2017. 5. Of?ce ofthe Surgeon General: National Action Alliance for Suicide Prevention. 2012 national strategy for suicide prevention: goals and objectives for action. Washington, DC: US Department of Health and Human Services. Of?ce of the Surgeon General: 2012. . surgeongeneral. go vi 1 i braiyr? rep prevention! in l- 6. Zalsman G, Hawton K, D, et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 20] 6:3:646?59. 7. Torguson K, O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington, DC: American Foundation for Suicide Prevention; 2017. 3. Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders Washington, DC: American Association: 2013. 10. Caine ED, Reed .1, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017. Epub December 20, 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package ot?policies, programs, and practice. Atlanta, GA: US Department of Health and Human Services, 2017. l3. Milner A, Svcticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int] Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona, Colorado. Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan, Minnesota, New Hampshire, New Jersey, New Mexico. New 1l?orlc, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Summaryl What is already known about this topic? In 2016, nearly 45,000 people died by suicide in the United States. Mental health conditions can contribute to suicide. What is added by this report? During 1999?2016, suicide rates increased in nearly every state, including >30% increases in 25 states. 2015 data from 2? states indicate 54% of suicide decedents were not known to have mental health conditions. Other contributors included relationship, substance use, health, and jobi??nancial problems, among others. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in CDC's Technical Package for Suicide Prevention, can help reach the national goal of reducing the annual suicide rate 20% by 2025. TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged :10 years with and without known mental health conditions - National Death Reporting System, 2? states) 2015 Known mental No known mental Total health conditiont health condition Adjusted Characteristic (N 20,445) (n 9,40?) (n 11,039) F-value 0R5 (95% Cl) (95% Cl) Sex Male 15,?02 5,459 (58.8) 9,233 (83.5) 0.01 2.3 Female 4,744 (23.2) 2,938 (31.2) 1,805 (15.4) 0.01 0.4 Ase 10?24 2,804 1,211 (12.9) 1.593 (14.4) 0.01 1.1 25?44 5,455 (31.5) 3.035 (32.3) 3,420 (31.0) 0.05 0.9 45?54 3,820 (40.5) 3,898 (35.3) 0.01 0.8 255 3.458 1,340 (14.2) 2,128 (19.3) 0.01 1.4 (1.3-1.5) White, non-Hispanic 1?,102 (83.5) 8.155 (85.8) 8,93? (81.0) 0.01 0.5 Black, non-Hispanic 1,22?50) 411 51? GA) 0.01 American IndianIAlaska Native, non?Hispanic 3?8 (1.8) 112 (1.2) 255 (2.4) 0.01 2.0 Asian. non-Hispanic 575 (2.8) 235 (2.5) 341 (3.1) 0.05 1.2 (1.1-1.5) Hispanic 1.095 (5.4) 483 (4.9) 533 0.05 1.2 Other 55 (0.3) 21 (0.2) 45 (0.4) 0.05 1.8 Extended demograghics Ever served in military"r 3,429 1,354 (15.3) 2,0?5 (20.1) 0.01 1.4 1.1 (1.0?1 .1) Homeless 240 (1.2) 104 (1.1) 135 (1.3) 1.1 1.2 (0.9-1.5) incident Type Single suicide 20,053 (98.2) 9,318 (99.1) 10,?45 0.01 0.3 0.4 Homicide followed by suicide 319 (1.5) 54 255 (2.3) 0.01 3.5 2.9 Multiple suicides 54 (0.3) 25 (0.3) 39 (0.4) 1.3 1.5 Method Firearm 9.909 (48.5) 3.821 (40.5) 5.088 (55.3) 0.01 1.8 1.5 5,90? (28.9) 2,940 (31.3) 2,95? (25.9) 0.01 0.8 0.8 Poisoning 3,003 1,851 (19.8) 1,142 (10.4) 0.01 0.5 0.5 Substance class causing death?? Other over-the-counter) 1,021 (34.0) 555 (35.8) 355 (31.1) 0.01 0.8 0.9 Opioids 944 (31.4) 508 335 (29.4) 0.9 0.9 (0.8?1 .1) Antidepressants 800 (25.5) 544 (34.5) 155 0.01 0.3 0.3 Benzodiazepines 524 (20.8) 458 (25.1) 155 0.01 0.5 0.5 219 (7.3) 195 (10.5) 24 (2.1) 0.01 0.2 0.2 Other 1,595 780 (8.3) 815 0.05 0.9 0.9 Toxicology Results An).r toxicology testing 5,558 (T08) 5,559 (50.3) 0.01 0.5 Positive for 21 substance?? 9,913 5,192 4,?21 0.01 0.8 Substance Alcohol Tested 10,950 (53.5) 5.409 5,541 (50.2) 0.01 0.8 Positive 4,442 (40.5) 2,115 (39.1) 2,32? (42.0) 0.01 1.1 1.2 Opioids Tested I 8,554 4.258 4,295 0.01 0.8 0.8 Positive I 2,2?9 1,238 1,041 0.01 i 0.8 0.9 Benzodiazepines Tested 8,124 4,225 (44.9) 3,898 (35.3) 0.01 0.2? Positive 2,454 (30.3} 1.539 (38.8) 325 (21.2) 0.01 0.4 0.5 Cocaine Tested {39.0) 3,855 (41.1) 4,112 0.01 0.9 0.9 Positive 499 215 (5.5) 283 (6.9) 0.05 1.2 1.2 Amphetamines Tested 1515 3.595 (39.3) 3,919 (35.5) 0.01 0.9 (0.5-0.9) 0.9 Positive 735 3?5 (10.2) 360 (9.2) 0.9 1.0 (0.3?1 .1) Marijuana Tested 6,559 (32.1} 3,127 (33.2) 3,442 (31.2) 0.01 0.9 0.9 Positive 1,4?1 {22.4) T10 T51 (22.1) -?11'i 1.0 0.9 (0.3-1.0) Antidepressants Tested 5,425 (26.5) 3,103 (33.0) 2,322 (21.0) 0.01 0.5 0.6 Positive 2,214 {40.8) 1,?35 (55.9) 479 (20.6) 0.01 0.2 0.2 Abbreviation: CI con?dence interval; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decadent had been identified as having a current diagnosis of mental health condition in ooronerfmedical examiner or law enforcement reports. ?4 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. YT Denominator is decedents aged 218 years with reported military service status. ?t Denominator is decedents who died by poisoning, including overdose. Tm Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Not significant. TABLE 2. Circumstances preceding suicide among decedents aged 21.0 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental health No known mental health condition?f condition P- Adjusted Characteristic Total no. no. value DR5 (95% Cl) Cl) Suicide with known circumstances 18,754 (91.8) 9.40? (100) 9.35? {84.5) s0.01 Mental health Any Current Diagnosed Mental Health Condition? (T52) Anxiety disorder 1,5?9 {15.3) Bipolar disorder 1,431 {15.2) Schizophrenia 509 (5.4) PTSD 424 (4.5) ADDIADHD 228 Unknown ?80 Current depressed mood 3.982 {42.1} 3,0?8 {32.9} ?0.01 Substance problems Any Current substance problem 5.319 {28.3} 2.9?8 {31.8} 2.343 {25.0} ?0.01 Alcohol problem 3.288 1.882 {19.8} 1,408 {15.0} ?0.01 Other substance problem 3.084 {18.4} 1.?88 {18.8} 1,318 {14.1} ?0.01 Treatment Current mental healthisubstanca abuse treatment 5.141 {54.0} 84 <0.01 0.01 {0.01?0.01} 0.01 {0.01?0.01} Ever treated for mental problem (35.8} 8.323 394 ?0.01 0.02 {0.02?0.02} 0.02 {0.02?0.03} Relationship problemsfloss Any relationship problemrloss ?.948 {42.4} 3.?28 {39.8} 4,222 {45.1} ?0.01 1.3 {1 1.3 {1 Intimate partner problem 5.098 2.2?0 {24.1} 2,828 {30.2} s0.01 1.4 1.4 {1 Perpetrator of interpersonal violence in past month 414 131 283 ?0.01 2.2 {1 2.0 Victim of interpersonal violence in past month 84 53 31 ?0.05 0.8 0.8 Family relationship problem 1,8?1 873 ?98 0.9 1.0 Other relationship problem {nonintimate} 403 202 201 ?11 1.0 1.1 Argument or con?ict {not specified) 2.914 {15.5} 1.2?8 {13.8} 1.838 {0.01 1.3 {1 1.4 {1 Death of a loved one {any} 1,49? 828 8?1 ?0.01 0.8 0.9 Nonsuicide death 1,181 84? 534 ?0.01 0.8 0.9 Suicide of family_or friend 3?9 21? 182 {1 ?0.01 0.8 Other life stressors Any life stressor 9.?43 {51.9} 4.8?5 5.088 {54.2} ?0.01 1.2 {1 1.1 {1 Recent criminal legal problem 1,588 588 1,002 ?0.01 1.8 Other legal problem ?48 3?8 3?0 ?11 1.0 1.0 Physical health problem 4.1?9 {22.3} 2.012 {21.4} 2.18? {23.2} ?0.01 1.1 {1 1.0 {1 JobiFinancial problem?? 2941 {18.2} 1530 {18.8} 1411 {15.8} ?0.05 0.9 0.9 Eviction or loss of home ?22 31? 405 ?0.01 1.3 1.4 {1 School problem'f?IT 182 {19.9} ?0 92 {21.9} ?Tt 1.3 1.3 Recent release from an institutionm 1,412 941 {10.2} 4?1 ?0.01 0.5 0.5 JailiPrisoniDetention facility 203 (14.4} 82 121 <0.01 3.8 4.5 Hospital 51? (38.8} 311 {33.0} 208 ?0.01 1.8 {1 1.3 {1 hospitali'institution 489 (33.2} 439 30 ?0.01 0.1 0.1 Other {includes alcohoIISA treatment facilities} 223 {15.8} 109 {11.8} 114 {24.2} ?0.01 2.4 2.5 Recent or Impending Crisis Crisis within past or upcoming 2 weeks?1 5.525 {29.4} 2.444 {28.0} 3,081 {32.9} ?0.01 1.4 {1 1.4 Intimate partner problem crisis 1988 {35.8} 854 {34.9} 1114 {38.2} ?11 1.1 1.1 Physical health problem crisis ?39 (13.4} 315 {12.9} 424 {13.8} ?1't 1.1 1.0 Criminal legal problem crisis 821 {11.2} 203 418 {13.8} ?0.01 1.8 {1 .3?19} problem crisis 430 212 218 ?0.05 0.8 0.9 Job problem crisis 354 191 183 <0.01 Suicide eventfhistory Left a note 8.488 {34.5} 3,182 {33.8} 3.288 {35.1} ?1't 1.1 {1 1.2 {1 Disclosed suicide intent 4.405 {23.5} 2,308 {24.5} 2,099 {22.4} ?0.01 0.9 0.9 History of ideation 5.990 {31.9} 3.838 {40.8} 2,152 {23.0} ?0.01 0.4 0.4 History of attempts 3.?32 {19.9} {29.4} 982 {10.3} ?0.01 0.3 0.3 Abbreviations: attention de?cit disorder-{attention deficit hyperactivity disorder; OR odds ratio; PTSD postlraumatic stress disorder; SA substance abuse. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South lCarolina, Utah, Vermont, Virginia, and Wisconsin. Decadent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. *3 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OR with 95% l[Sis after controlling for age, sex, race, and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. YT Not significant. Denominator is decedents aged 3:18 years. Denominator is decedents aged 10?18 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonly occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Comment Ll. Please replace ?mental health problems" with ?mental healih conditions" throughout. By de?nition in methods. these are medical 1' conditions. Calling their] ?problems If trivializcs them and adds further stigma I Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah Stone Thomas R. Simon Katherine A. Fowler. Scott Kegltr Kcmina Yuan Kristin Holland Asha Z. Pth: Alex E. Crosby- r1' 1 because a person should be able to cope with their problems in our culture. Most are conditions that can be medically treated. 1 Also, MMWR is a the scienti?c basis for the fact sheet, and all but the summary box Abstract lntroductmn Suicide rates in the United States have risen nearly 31] 0 since 1999 and mental health 'ondition . lire iust one factor contributing to suicide. Examining state-level trends in suicide and; 51101119 be ?Title? in SCi?nti?E {Millim- the multiple circumstances contributing to it can inform comprehensive slate suicide prevention planning. "a 1 Happy to discuss w1lh authorl {hit Methods: Trends in agcladjustcd suicide rates among persons aged 310 years by state and sex, across six ll) ,1 6'1 consecative 3-year periods (1999?2016}, were assessed using data From the National Vital Statistics System ?a for 50 states and the District of Columbia Data from the National Violent Death Reporting System, Comment From Michael covering 27 states in 2015 were used to examine contributing circumstances among with and Iadcmarco: without known mental health Wproblomh. Stigmatizing Ianguaga I don?t want to rock the boat so late int eh process. Are these mental health "problems?" I want to be sure we Results: During 1999?2016, suicide rates increased significantly in 44 states 25 states expeiiencing increases Rates increased signi?cantly among males and females in 34 and 43 states respectively. [n 2015 a total of549? of decedents In 2? states did not have a known mental health go__i;i_t1i_t_igii- emblem. . . . . Among decedeuts with available information ran-circumstances? available several possible contributing are 1101111511151 . . . I language. How about "conditions?? Circumstances Were common iinonL all [loo-ever. ere si ntiantl more likel amon . l" l? i? This occurs throughout the than among those with mental health . manuscript. In Methods, It seems the data source will support the those without a known mental health conditionproblems including relationship problem loss (45 1?41 versus 39. life sircssors {54 2% versur. and crises {32 9% versua 2o. of the word conditions and ?1 Lnit-:ignoses. Comment Mental health condition different than relationship across?groom Conclusions: Suicide rates increased signi?cantly across most states during 1999? 201 6 Various 1 Circumstances contnbutcd to suicides among peisons with and without known mental health problems. - . .. - Lproblcig, even if both can be devastating. Comment L3. Find confusing. 1 Implications for Public Health Practice: States can use a comprehenswe evidence based public health approach to prevent soicide risk before it occurs, identify and support persons at risk, prevent realtempis Comment Please go through rest of report and change. and help friends and family in the aftermath of a su1c1dc Introduction In 2016, nearly 45,000 suicides (15.61? 100,000 population occurred in the United States among persons aged 1:11] years ['11 From 1999 to EMS, suicide rates increased among both sexes. all raciali'ethnic groups. and all urbanization levels Suicide is the 10th leading cause ofdeath and is one ofjust three leading causes that are increasing {i 4) in addition rates of emergency department visits for suicides and self-harm injuiies? cost the nation approximately 569 billion in direct medical and work loss i costs The National Strategy for Suicide Prevention (5) calls for a public health approach to su1c1dc Il prevention with cli'orts spanning multiple levels (I 1: Individual Familyfrelationship, community; and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor but g' rather is determined by multiple factors. Despite NSSP guidance, suicide prevention largely focuses on i identifying wields} persons providing treatment for mental health Comment IKCU: L3. Because used only 2 more times. so est 5 ell throu rhout. Comment L3. To make it easier for reader, please define abbreviations for states in this table. Some people might not have all 50 abbreviations in their headg. reatlempts In addition to mental health conditionsproblenss and prior wit itlc attempts. other circumstances social and economic problems. access to lethal means te.g.. substances. firearms) among persons at risk. and poor coping and problem-solving skills Expanded awareness of these additional circumstances; contributing to suicide risk and action to address them can help reach the national goal established by the American Foundation for Suicide Prevention and the National fiction Alliance of Suicide Prevent of reducing suicide rates by 20% by 2925 (it. To assist ?l states in achieving this goal. CDC analyzed state-speci?c trends in suicide rates and assessed the multiple ll contributing factors to suicide; this report presonts options for multilevel comprehensive suicide prevention i i based on the best available evidence. Methods Suicide rates were analyzed for persons aged 1:10 years only. as determining suicidal intent in younger children can be dif?cult till. Age-Specific suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (International Classification of Diseases, Tenth Roi-'ist'on. codes U03). Age?specific population estimates were obtained from US. Census BurcauiNational Center for Health Statistics bridged?race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999?2016: [1999?2001; 2002?2004; 2005?2007; 2008?2010; 201 1?20] 3; and 2014? 2016}. Rate estimates were age?adjusted to the U.S. 2000 standard population and expressed per 100.000 persons per year. Ageaadjustcd suicide rate trends were modeled using the same 3?year data aggregates. employing weighted least-squares regression with inverse?variance weighting. Modeled rate trends are repented in terms of average annual percentage changes Characteristics of persons aged 3210 years who died by suicide: with land ?withouihnown itieittal? liealih problems. and the circumstances surrounding the suicides were compared in the 2? states* with complete data participating in National Violent Death Reporting System in 5. defines mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Memo! Disorders, i?h Edition with the exception of' alcohol and other substance use disorders. which are captured separately in aggregates data from three primary data sources: death certi?cates. coronerfmedical examiner reports (including toxicology]. and law enforcement reports. Decedents with and without known mental health problems were compared using Chi?square tests. Logistic regression analyses estimated adjusted odds ratios {aORsl with 95% confidence- intervals tCIs]. controlling for age group. sex. and raceiethnicity. Results The most recent overall suicide rates {re resenting 2014?2?16) varied fourfold. from I19 (DC) to 29.2 (Montana) Per 100%) persons Per tear Across the study period. rates increased in all states except Nevada (where the rate was consistently high throughout the study period]. with absolute increases ranging item 0.8 per {Delaware} to ELI [Wyoming]. Percentage increases in rates ranged From 5.9% (Delaware) to 515% (North Dakota]. with increases observed in 25 states [Supplementary Table: [Figure?t). Modeled suicide rate trends indicated signi?cant increases in 44 states. among males {34 states] and females [43 states), as well as for the United States overall (Supplementary Table; l. Nationally. the model-estimated AAPC for the overall suicide rate was an increase of By sex. estimated national rate trends Further indicated signi?cant AAPC increases and females (Supplementary Table; for males 1.1% Suicide decedents without known mental health problems (11,039) were compared with those with known mental health problems (9,407) in 27 states. Whereas all decedents were predominately male (Table l) and non-Hispanic white those without known mental health problems, relative to those with mental health problems, were more likely to be male (83.6% versus 68.8%; odds ratio 95% and belong to a racialfethnic minority (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of peipetrating homicide-suicide 2.9, 95% CI Among adult decedents aged 3:18 years, 20.1% of those without known mental health problems and 15.3% ofthose with mental health problems had ever served_uj or were currently Ill .54 milieu-y. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health problems, decedents without known mental health problems, relative to those with known mental health problems, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained significant in the adjusted models. 'I'tcx itoltgg, testing tx- :1 likely to bu- rut-riotoutl withoutknown mental health problems Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall (30R 0.3, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (n 9,407) and approximately 85% of those without known mental health problems (it 9,357) in 27 states (Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% CI than were persons with known mental health problems. Whereas two thirds of decedents with known mental health problems had a history of mental health or substance use treatment just over half were in arcane-treatment 14$ rut itsth, Decedents without known mental health problems had a. signi?cantly higher likelihood of any relationship problenu?loss than did those with known mental health problems speci?cally intimate partner problems (30.2% versus arguments/conflicts (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than were those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminalflegal problems (10.7% versus or evictioniloss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained significant in the adjusted models. Among all. persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobi??nancial problems were commonly experienced among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.3%, respectively). Decedents without known mental health problems had significantly lower odds of recent release from any institution (a0R=0.5, 95% but those who were recently released were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcohol/substance treatment) 2.5 95% C1 than were those with a known mental health problem. Among decedents with known mental health problems who were recently released from an institution 46.7% were released from facilities. Decedents without known mental health problems were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems [40.8% and 29.4%. respectively). Suicide intent was disclosed by 22.4% and ofpersons without and with known mental health problems. respectively. Conclusions and Comments During l999 2016. suicide rates increased significantly in 44 states. and 25 states experienced increases >3tl9xli. Rates increased signi?cantly among males in 34 states. and females in 43 states. This ?nding is consistent with prior research that indicated a decreasing gender gap in male-female suicide rates during 1999?2014 (3). Additional research into the speci?c causes of these trends is needed. Data from the 2? . states participating in provide insight into circumstances surrounding suicide and can Ii 5' help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however. suicide prevenliun is often oriented toward identi?cation of suicidal persons, treatment of mental health problems. and prevention of reattempts. This study found that approximately half of suicide decedents in did not have a known mental health problem. indicating that additional focus on nonmental health factors further upstream could provide important information for a public health approach ill]. Those without a known mental health problem suffered more from relationship problems and other life stressors such as matters. cvictioniloss of home, and recent or impending crises. Similarly, persons with mental health problems olien experienced relationship problems and other life stressors such as jobi?linancial or physical health problems. (These ?ndings point to the need to help persons both manage and prevent the conditions associated with mental health ggnditionspeelelems in the first place. and to support persons with known mental health oondilionsprobleens to decrease their risk for poor; . outcomes suicide _y_v_itl1_ mentaLl-lp?l?l conditionspf?b-lrepjihad?pj history of treatment for any mental health or substance use or both. with approximately half in treatment when they died This ?nding suggests the need for additional safety supports. including broader implementation of? affordable and effective treatment modalities, such as doctor-patient collaborative care 1' models and proven cognitive-behavioral therapies. In addition. increased access to behavioral health providers in areas is needed. as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports (cg. housing stabilization policies. household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent relationship problems. capecially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible. emotional, and social support: and identifying and better supporting persons at risk military veterans. persons with physicalimental health problems} U2). Other strategies include creating protective environments reducing access to lethal means anton persons at promote helpeseeking. casing transitions into and out of work for persons with mental health problems and other life challenges]. supporting family and friends after a suicide, and assuring the media Follow safe reporting guidelines Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention (ill). The findings in this report are subject to at least three limitations. First. in tlte state-level analysis. rankings for four states (Maryland, Massachusetts. Rhoda island. and Utah) might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reponed suicide rates downward) or decreased percentages of such deaths over time {potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 2? states included represent 49.6% of the population govifacesr'tablescn'icesi?j s?'page sfproductviewahtml l. Finally. Comment 1KCU: L2. Has this been reviewed by a Not sure what is meant by "prevent the conditions". [Comm-3m Michael lademarcn: L3. MMWR, of self-in?icted abstraetors of data are limited to information contained in investigative reports. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of?kin often identify greater attributions to mental disorders however. many methodological variations across studies exist (14). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless. the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies. Programs. and Practices (12) to better understand and manage t-heiesuicide in their communitiespreblem. prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgments Robert Anderson. Holly Hedegaard. Margaret Warner. Division of Vital Statistics. National Center for Health Statistics. CDC. Conflict of Interest No conflicts ot'interest were reported. 'Division ofViolenec Prevention, National Center for Injury Prevention and Control. CDC: 1Division of Analysis. Research. and Practice Integration. National Center for Injury Prevention and Control, CDC. Corresnonding author: Deborah M. Stone. dstone3?cdegom "rm-4884942- References I. CDC. Web-based Injtu'y Statistics Query and Reporting System Atlanta. GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ. Crosby AE. Jack SPD. Haileyesus T. Kresnow-Sedacca MJ. Suicide trends among and within urbanization levels by sex. raeefethnicity. age group. and mechanism of death?United States. 2001?2015. MMWR Surveill Summ 88-18). 3. Curtin SC. Warner M. Hedegaard H. Increase in suicide in the United States. 1999?2014. NCHS data brief no 24]. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2016. .pdf 4. Kochanek K. Murphy S, Xu J. Arias E. Mortality in the United States. 2016. data briefno 293. [-l'yattsville. MD: US Depaitment of Health and Human Services. CDC. National Center for Health Statistics: 2017'. 5. Of?ce of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 national strategy for suicide prevention: goals and objectives for action. Washington. DC: US Department of Health and Human Services. Of?ce of the Surgeon General; 2012. 6. Zalsman G. Hawton K, Wasserman D. et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 2016;3z?46?59. Torguson K. O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; 2017. 3. Crepeau-Hobson .F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders 5). Washington, DC: American Association; 2013. 10. Caine ED. Reed Hindman J. Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Prev 2017. Epub December 20. 2017'. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva. Switzerland: World Health Organization: 2012. 12. Stone DM. Holland KM. Bartholovv BN. Crosby AE. Davis SP. Wilkins N. Preventing suicide: a technical package of policies. programs, and practice. Atlanta. GA: US Department of Health and Human Services. 2017. l3. Milner A. Sveticic J. De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona. Colorado, Connecticut. Georgia, Hawaii. Kansas. Kentucky. Maine, Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode island. South Carolina. Utah. Vennont. Virginia. and Wisconsin. Summary What is already known about this topic? In 2016. nearly 45.000 deaths were caused by suicide in the United States. Mental health conditions often contribute to suicide. What is added by this report? During 19994015. suicide rates increased in nearly every state. including >30% increases in 25 stateszleetal 5 data from the {27 states) indicate 54% of suicide decedents were not known to have mental health conditionssosh?ppeeiems. Other contributors included relationship. substance use. health. and jobt?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in CDC's Technical Package for Suicide Prevention. can help reach the national goal of reducing suicide rates 20% by 2025. TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged :10 years with and without known mental health problems National lliiolent Death Reporting System, 2? states) 2015 Known mental No known mental Total health printablem?r health problem Adjusted Characteristlc (N 20,445) (n 9,40?) (n 11,039) F-value 0R5 (95% Cl) (95% Cl) Sex Male 15,?02 5,459 (58.8) 9,233 (83.5) 0.01 2.3 Female 4,744 (23.2) 2,938 (31.2) 1,805 (15.4) 0.01 0.4 Ase twat? 10?24 2,804 1,211 (12.9) 1.593 (14.4) 0.01 1.1 25?44 5,455 (31.5) 3.035 (32.3) 3,420 (31.0) 0.05 0.9 45?54 3,820 (40.5) 3,898 (35.3) 0.01 0.8 255 3.458 1,340 (14.2) 2,128 (19.3) 0.01 1.4 (1.3-1.5) White, non-Hispanic 1?,102 (83.5) 8.155 (85.8) 8,93? (81.0) 0.01 0.5 Black, non-Hispanic 1,22?50) 411 51? GA) 0.01 American IndianIAlaska Native, non?Hispanic 3?8 (1.8) 112 (1.2) 255 (2.4) 0.01 2.0 Asian. non-Hispanic 575 (2.8) 235 (2.5) 341 (3.1) 0.05 1.2 (1.1-1.5) Hispanic 1.095 (5.4) 483 (4.9) 533 0.05 1.2 Other 55 (0.3) 21 (0.2) 45 (0.4) 0.05 1.8 Extended demographics Ever served in military"r 3,429 1,354 (15.3) 2,0?5 (20.1) 0.01 1.4 1.1 (1.0?1 .1) Homeless 240 (1.2) 104 (1.1) 135 (1.3) 1.1 1.2 (0.9-1.5) incident Type Single suicide 20,053 (98.2) 9,318 (99.1) 10,?45 0.01 0.3 0.4 Homicide followed by suicide 319 (1.5) 54 255 (2.3) 0.01 3.5 2.9 Multiple suicides 54 (0.3) 25 (0.3) 39 (0.4) 1.3 1.5 Method Firearm 9.909 (48.5) 3.821 (40.5) 5.088 (55.3) 0.01 1.8 1.5 5,90? (28.9) 2,940 (31.3) 2,95? (25.9) 0.01 0.8 0.8 Poisoning 3,003 1,851 (19.8) 1,142 (10.4) 0.01 0.5 0.5 Substance class causing death?? Other over-the-counter) 1,021 (34.0) 555 (35.8) 355 (31.1) 0.01 0.8 0.9 Opioids 944 (31.4) 508 335 (29.4) 0.9 0.9 (0.8?1 .1) Antidepressants 800 (25.5) 544 (34.5) 155 0.01 0.3 0.3 Benzodlazepines 524 (20.8) 458 (25.1) 155 0.01 0.5 0.5 219 (7.3) 195 (10.5) 24 (2.1) 0.01 0.2 0.2 Other 1,595 780 (8.3) 815 0.05 0.9 0.9 Toxicology Results An).r toxicology testing 5,558 (T08) 5,559 (50.3) 0.01 0.5 Positive for at substance?? 9,913 5,192 4,?21 0.01 0.8 Substance Alcohol Tested 10,950 (53.5) 5.409 5,541 (50.2) 0.01 0.8 Positive 4,442 (40.5) 2,115 (39.1) 2,32? (42.0) 0.01 1.1 1.2 Opioids Tested I 8,554 4.258 4,295 0.01 0.8 0.8 Positive 2,2?9 1,238 1,041 0.01 0.8 0.9 Benzodiazepines Tested 3,124 4,226 (44.9) 3,898 [35.3) 0.01 Positive 2,464 (30.3} 1.639 (38.8) 325 (21.2) 0.01 0.4 0.5 Cocaine Tested (39.0) 3,866 (41.1) 4,112 (32.2) 0.01 0.9 0.9 Positive 499 215 283 (6.9) 0.05 1.2 1.2 Amphetamines Tested 1615 {312} 3.696 (39.3) 3,919 (35.5) 0.01 0.9 (0.6-0.9) 0.9 Positive 75613.?) 3?6 (10.2) 360 (9.2) 0.9 1.0 (03?1 .1) Marijuana Tested 6,559 (32.1} 3,127 {33.2) 3,442 [31.2) 0.01 0.9 0.9 Positive 1,4?1 {22.4) T10 ?61 (22.1) 1.0 0.9 (0.6-1.0) Antidepressants Tested 5,425 {25.5) 3,103 (33.0) 2,322 (21.0) 0.01 0.5 0.6 Positive 2,214 (40.8} 1,735 (55.9) 479 [20.6) 0.01 0.2 0.2 Abbreviation: CI con?dence interval; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decadent had been identified as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. it OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race, and ethnicity. Known mental health problem was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. it Denominator is decedents aged :18 years with reported military service status. *6 Denominator is decedents who died by poisoning, including overdose. Til Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged :10 years with and without known mental health problems National Violent Death Reporting System, 27 states,* 2015 Known mental health No known mental health problem1 problem P- Adjusted (95% Characteristic Total no. ?lls) no. value OR5 {95% Cl) Cl) Suicide with known circumstances 13364 (91.8) 9,40?Ir {100) 9.35? (84.8) <0.01 Mental health Any Current Diagnosed Mental Health Problem? 1075 Anxiety disorder 1,5?9?69) Bipolar disorder 1,431 {15.2) Schizophrenia 509 PTSD 424 226 (2.4) Unknown ?60 (3.1) Current depressed mood 3,962 (42.1) 3,0?6 (32.9) ?0.01 Substance problems Any Current substance problem 5,319 (23.3) 2.9?6 (31.6) 2,343 (25.0) <0.01 Alcohol problem 3,268 1,362 (19.3) 1,406 (15.0) <0.01 0.7 Other substance problem 3.034 (16.4) 1,?63 (13.3) 1,316 (14.1) {0.01 Treatment Current mental healthisubstance abuse treatment 5,141 50?? (54.0) 64 (0.01 0.01 (0.01-0.01) 0.01 (0.01?0.01) Ever treated for mental problem (35.3) 6,323 394 (4.2) <0.01 0.02 (0.03-0.02) 0.02 (0.02?0.03) Relationship problemsiloss Any relationship problemfloss ?.948 (42.4) 3,?26 (39.5) 4,222 (45.1) ?0.01 1.3 1.3 (1 Intimate partner problem 5,098 2,2?0 (24.1) 2,328 (30.2) <0.01 1.4 1.4 (1 Perpetrator of interpersonal violence in past month 414 (2.2) 131 (1.4) 283 (3.0) 40.01 2.2 2.0 (1.6-2.4) Victim of interpersonal violence in past month 84 (0.4) 53 (0.6) 31 (0.3) <0.05 0.6 (0 4?0.9) 0.8 Family relationship problem 1,6?1 (3.9) 3?3 (9.3) 793 (8.5) 0.9 1.0 Other relationship problem (nonintimate) 403 (2.1) 202 (2.1) 201 (2.1) 1.0 1.1 Argument or con?ict (not specified) 2,914 (15.5) 1,2?3 (13.6) 1,636 (1 <0.01 1.3 1.4 (1 Death of a loved one (any) 1,49? (8.0) 326 (3.3) 6?1 <0.01 0.3 0.9 (0.3-0.9) Nonsuicide death 1,181 (6.3) 64? (6.9) 534 <0.01 0.3 0.9 Suicide of family or friend 379 (2.0) 21? (2.3) 162 <0.01 0.8 Other life stressors Any life stressor 9,?43 (51.9) 4,6?5 5,068 (54.2) <0.01 1 2 1 1 (1 Recent criminal legal problem 1,588 (8.5) 586 (6.2) 1,002 <0.01 1 3 (1.6-2.0) 1 (1.5-1.9) Other legal problem ?48 (4.0) 3?8 (4.0) 370 (4.0) 1 0 (0.3?1 .1) 1.0 (0.9-1.2) Physical health problem 4,1?9 (22.3) 2,012 (21.4) 2,16? (23.2) <0.01 1 1 1.0 (1 JobiFinancial problem?? 2941 (16.2) 1530 (16.3) 1411 (15.6) ??0.05 0 9 0.9 Eviction or loss of home ?22 (3.3) 31? (3.4) 405 (4.3) <0.01 1.3 1.4 (1 School problem?" 162 (19.9) ?0 (1 92 (21.9) 1.3 1.3 (0.9-1.9) Recent release from an 1,412 941 (10.2) 4?1 (5.1) <0.01 0.5 0.5 (0.4-0.5) JailiPrisoniDetention facility 203 (14.4) 32 121 ??0.01 3.6 4.5 Hospital 51? (36.6) 311 (33.0) 206 <0.01 1.6 1.3 (1 hospitaliinstitution 469 (33.2) 439 30 (6.4) <0.01 0.1 0.1 (0.1411) Other (includes alcohoIISA treatment facilities) 223 (15.3) 109 (11.6) 114 (24.2) <0.01 2.4 2.5 (1.3-3.3) Recent or Impending Crisis Crisis within past or upcoming 2 weekst?r?r 5,525 (29.4) 2,444 (26.0) 3,031 (32.9) ??0.01 1.4 1.4 (1 Intimate partner problem crisis 1968 (35.6) 354 (34.9) 1114 (36.2) 1.1 1.1 Physical health problem crisis "(39 (13.4) 315 (12.9) 424 (13.8) 1.1 1.0 (0.3-1.2) Criminal legal problem crisis 621 (11.2) 203 (8.3) 418 (13.6) <0.01 1.6 (1 Family relationship problem crisis 430 (7.3) 212 213 <0.05 0.3 0.9 Job problem crisis 354 (6.4) 191 163 (5.3) <0.01 0.7 Suicide eventfhistory Left a note 6,468 (34.5) 3,132 (33.8) 3,286 (35.1) 1.1 (1.0?1 .1) 1.2 (1 .1?12) Disclosed suicide intent 4.405 (23.5) 2,306 (24.5) 2,099 (22.4) <0.01 0.9 0.9 (0.3-0.9) History of ideation 5,990 (31.9) 3,833 (40.3) 2,152 (23.0) <0.01 0.4 0.4 History of attempts 3,?32 (19.9) Q94) 962 (10.3) <0.01 01303?03) 0.3 Abbreviations: ADDFADHD attention de?cit disorderi'attention deficit hyperactivity disorder; OR odds ratio; PTSD posttraumatic stress disorder; SA substance abuse. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health problem was the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. it Not significant. ?t Denominator is decedents aged :18 years. Tm Denominator is decedents aged 10?13 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonlv occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler. Scott R. Kegler. Kenning Yuan. Kristin M. Holland. Asha Z. Hey-Stephenson. Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National 1'v?ital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 20l5, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states. with 25 states experiencing increases 80%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. in 2015, a total of 54% of decedents in 27 states did not have a known mental health problem. Among decedents with infonnation on circumstances available. several circumstances were signi?cantly more likely among those without a known mental health problems than among those with mental health problems, including relationship problems/?loss {45.1% versus life stressors (54.2% versus and crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. introduction In 2016, nearly 45,000 suicides (15.6fl 00,000 population [age-adjustedD occurred in the United States among persons aged 210 years From 1999 to 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide is the 10th leading cause ofdeath and is one of just three leading causes that are increasing .4). In addition, rates of emergency depaitment visits for nonfatal self-harm, a key risk factor for suicide. increased 42% item 2001 to 2016 (1). Together, suicides and self- harm injuries cost the nation approximately $69 billion in direct medical and work less costs The National Strategy for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite NSSP guidance. suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems, and preventing reattempts 6). In addition to mental health problems and prior attempts. other circumstances contributing to suicide include social and economic problems, access to lethal means (cg, substances, ?rearms) among persons at risk. and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal established by the American Foundation for Suicide Prevention and the National Action Alliance of Suicide Prevent ofreducing suicide rates by 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-specific trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged I310 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (International Cinssificntion of Diseases, Tenth Revision. underlying-cause-ot? death codes Y87.0, U03). Age-speci?c population estimates were obtained from US. Census BureauiNational Center for Health Statistics bridged?race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999?2016 (1999?2001; 2002?2004; 2005?2007; 2008?2010; 2011?2013; and 2014? 2016). Rate estimates were age-adjusted to the US. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, Employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics of persons aged 310 years who died by suicide, with and without known mental health problems, and the circumstances surrounding the suicides were compared in the 27 states* with complete data participating in National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Siniisiicni Menard of Manta! Disorders, Fifin Edition (9), with the exception of alcohol and other substance use disorders. which are captured separately in aggregates data from three primary data sources: death certi?cates, coroner/medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known mental health problems were compared using hi-square tests. Logistic regression analyses estimated adjusted odds ratios (aORs) with 95% con?dence intervals (Cls), controlling for age group, sex, and raceiethnieity. Results The most recent overall suicide rates (representing 2014?2016) varied fourfold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from 0.8 per 100,000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases 380% observed in 25 states (Supplementaiy Table: (Figure Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females {43 states), as well as for the United States overall (Supplementary Table; Nationally, the model-estimated AAPC for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated significant AAFC increases for males (1 and females (Supplementary Table; Suicide decedents without known mental health problems (11,039) were compared with those with known mental health problems (9.407) in 27 states. Whereas all decedents were predominately male (Table 1) and non-Hispanic white those without known mental health problems, relative to those with mental health problems, were more likely to be male (83.6% versus 68.8%; odds ratio 95% CI and belong to a racialr'ethnie minority (OR range U). Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide?suicide 2.9, 95% Among adult decedents aged 318 years. 20.l% of those without known mental health problems and of those with mental health problems had ever servedwere currently serving. in the U.S. military. I Whereas firearms were the most common method of suicide overall and among dcecdents with it and without mental health problems. decedents without known mental health problems, relative to those with known mental health problems. were more likely to die by ?rearm (55.3% versus 40.6%) and less it likely to die by (26.9% versus 31.3%} or poisoning {10.4% versus These differences remained signi?cant in the adjusted models. Deecdents without known mental health problems were less likely to receive toxicology testing. Among it those with toxicology results, without knoivn mental health problems were less likely to test positive for any substance overall taOR= 0.8. 95% Cl including opioids 0.90. 95% g: but were more likely to test positive for alcohol l.2, 95%. Cl 3 ll Information on circumstances surrounding suicide were available tor all deecdents with mental health problems (n 9,4ll'i') and approximately 85% of those without known mental health problems (rt 9,357) in 2? states {Table Persons without known mental health problems were less likely to have any substance use disorders 95% Cl than were persons with known mental health problems. Whereas two thirds of with knowln mental health problems had a history of mental health or ll substance use treatment just over half (54 13%) were in current treatment. it Deccdents without known mental health problems had a significantly higher likelihood of any relationship problemlloss {45. than did those with known mental health problems speci?cally intimate partner problems (30.2% versas argumentsi?eonllicts (17.5% versus and recently ii perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than were those with known mental health problems to have experienced any life .. ll Comment This is the section 1 found confusing. Comment So the ?rst number i is the of 5.l of decedents with no known mental health issue who were recently release. And. the second number is the of some of decedents with a known mental health issue who were recently released. correct?l As 1 said on the phone, I had to read several times to ?gure who Was who and where the percentages were coming from. Maybe something along the lines of? below: Deecdents without known mental health problems had signi ?eantly lower odds of having been recently released from any institution li.5. 95% Ct Among those recently released from an institution, dccedents without a known mental health issue when compared with those with a known mental health problem were signi?cantly more likely to have been released from a correctional facility versus hospital versus or other facility leg. alcohollsubstance treatment} stressors (54-2% versus 499%} such as problems ?09% versus or evictiont?loss of home versus and were more likely to have had a recent or impending [within the preceding or .1 upcoming 2 weeks. respectively] crisis (a current or acute event thought to contribute to the suicide} 32.9% versus All of these differences remained signi?cant in the adjusted models. Among all persons 5: with recent crises. intimate partner problems were the most common types and did not differ by group. i: Similarly, physical health problems and jobl?nancial problems were commonly experienced among both ll persons without mental health problems (23.2% and respectively} and those with mental health ll ll problems 4% and 16.8%, respectively]. [Become transitional:assailants nationalists anions any institution 0.5. 95% CI but those who were recently released were signi?cantly more likely to have been released from a correctional facility (19.59% versus [439% versus or other facility leg. treatment) taOR 2.5 95% CI than were those with a known mental health problem. Among with known mental health problems who were recently released from an institution 469% were released from facilities. Decedean without known mental health problems Were significantly less likely to have a history of suicidal ideation (23.0940 or prior suicide attempts I: compared with those with known mental health problems (40.8% and 29.4%. respectively). Suicide intent was disclosed by 23.4% and 24.5% of persons without and with known mental health problems, respectively. tiaUR 2.5 95% CI Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increases 360%. Rates increased signi?cantly among males in 34 states, and females in 43 states. This ?nding is consistent with prior research that indicated a decreasing gender gap in male-female suicide rates during 1999?2014 Additional research into the speci?c causes of these trends is needed. Data from the 27 states participating in provide important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identification of suicidal persons, treatment of mental health problems. and prevention of reattempts. This study found that approximately half of suicide decedents in did not have a known mental health problem, indicating that additional focus on nonmenta] health factors further upstream could provide important information for a public health approach (10). Those without a known mental health problem suffered more from relationship problems and other life stressors such as criminaldegal matters, evictioni?loss ofhome, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as job/financial or physical health problems. These ?ndings point to the need to help persons both manage and prevent the conditions associated with mental health problems in the ?rst place, and to support persons with known mental health problems to decrease their risk for poor outcomes {11' Two thirds of this group with mental health problems had a history of treatment for any mental health or substance use or both, with approximately halfin treatment when they died. This finding suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities, such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. in addition, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions U2). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household financial support): teaching coping and problem-solving skills to manage everyday stressors and prevent ?ature relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicalimental health problems} Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges}, strengthening access and delivery of care, supporting family and frieuds alter a suicide, and assuring the media follow safe reporting guidelines (12). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention (10). The ?ndings in this report are subject to at least three limitations. First, in the analysis, rankings for four states (Maryland, Massachusetts, Rhodc Island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent {potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of-kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown. and hence underreportcd by key informants. Nonetheless. the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policies. Programs. and Practices to better understand their suicide problem. prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgments Robert Anderson. Holly Hedegaard. Margaret Warner. Division of Vital Statistics. National Center for Health Statistics. CDC. Con?ict of Interest No con?icts of interest were reported. 'Division ol'Violence Prevention. National Center for Injury Prevention and Control. CDC: 1Division ofAnalysis. Research. and Practice Integration. National Center for I njury Prevention and Control. CDC . Corresponding author: Deborah M. Stone. dstonelit?gimdcgov. 770?438?3942. References l. CDC. Webwbased Injury Statistics Query and Reporting System Atlanta. GA: National Center for Injury Prevention and Control. Retrieved March 15. 2013. In. 2. Ivey?Stephenson AZ. Crosby AE. Jack SPD. Haileyesus T. Kresnow-Sedacca MI. Suicide trends among and within urbanization levels by sex. racet'ethnicity. age group. and mechanism of death?United States. 2001?2015. MMWR Surveill Summ 201?;6otNo. SS-IS). 3. Curtin SC. Warner M. Hedegaard H. Increase .in suicide in the United States. 1999? 2014. NCHS data brief no 241. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2016. .pdf 4. Kochanek K. Murphy S. Xu J. Arias E. Mortality in the United States. 2016. NCHS data briefno 293. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2017. 5. Of?ce ofthe Surgeon General: National Action Alliance for Suicide Prevention. 2012 national strategy for suicide prevention: goals and objectives for action. Washington. DC: US Department of Health and Human Services. Of?ce of the Surgeon General: 2012. . surgeongeneral. go vi 1 i braty/rcp -preventioitfii.t l- report.pdf 6. Zalsman G. Hawton K. D. et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 20] 6:3:646?59. 7. Torguson K. O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation for Suicide Prevention; 2017. 3. Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders Washington. DC: American Association: 2013. 10. Caine ED, Reed .1, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017. Epub December 20, 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package ot?policies, programs, and practice. Atlanta, GA: US Department of Health and Human Services, 2017. l3. Milner A, Svcticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int] Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona, Colorado. Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan, Minnesota, New Hampshire, New Jersey, New Mexico. New 1l?ork, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Summaryl 1What is already known about this topic? In 2016, nearly 45,000 deaths were caused by suicide in the United States. What is added by this report? During 1999?2016, suicide rates increased in nearly every state, including >30% increases in 25 states. Mental health problems often contribute to suicide; however, 2015 data from the National Violent Death Reporting System states) indicate 54% of suicide decedents were not known to have such problems. Other contributors included relationship, substance use, health, and jobl?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in CDC's Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025. . Comment HTM These need to be TABLE 1. Selected demographic and des:riptive characteristics of suicides among persons aged all] years with and without known mental health problems National . NIR Violent Death 2? states 2015 3m? 't (5?32? or new known (I tootuoted to explain], Total health Fin-ablerrtT health problem 1 I139 P-tralue 0H5 6 EB. [1131 2.3 2?2 3420 3120 American I non-His Ever served in Incident a suicide suicides 54 0.3 Firean?n 3909 43. ueclass Other . -l'h . 355 31.1 sanls . 156 13. 2 Substance detected?? 10 . Positive 4 442 40.6 1. 0.01 00?0. 0.7" 0.0.01 0.0 osnoe 40.8 1?35 55.9 47'9 0.01 0.2 02:02 02 0240.3 intenral: o. Alaska. Arizona. Colorado. Connecticut, Georgia. Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. Abbreviation: - New York. North Carolina, Ohio. Oklahoma. Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. Decadent had been identified as having a current diagnosis of mental health problem in ooronerirnedioal examiner or law enforcement reports. 5 OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 'l Logistic regression was used to estimate adjusted OR with 95% Die after controlling for age. sex. race. and ethnicity. Known mental health problem was used as the reference group. Decedents were aged 210 years. as per standard in the suicide prevention literature. Denominator is deoedents aged :18 years with reported military service status. $5 Denominator is decadents who died by poieoning. including overdose. Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. P- Adjusted on! {95% ti" Not signi?cant. TABLE 2. Circumstances preceding suicide among decedents aged 21D veers with and without known mental health problems National Violent Death Reporting System. 2? problem? problem no. no. slams 2015 Characteristic health 0T6 T5 1 1115 Problem 424 4.5 (Comment ditto a_bove Unknown ?60 (3.1) Current depressed mood 3,962 (42.1) 3,0?6 (32.9) ?0.01 Substance problems Any Current substance problem 5,319 (23.3) 2.9?6 (31.6) 2,343 (25.0) <0.01 Alcohol problem 3,268 1,362 (19.3) 1,406 (15.0) <0.01 0.7 Other substance problem 3.034 (16.4) 1,?63 (13.3) 1,316 (14.1) {0.01 Treatment Current mental healthisubstance abuse treatment 5,141 50?? (54.0) 64 (0.01 0.01 (0.01-0.01) 0.01 (0.01?0.01) Ever treated for mental problem (35.3) 6,323 394 (4.2) <0.01 0.02 (0.03-0.02) 0.02 (0.02?0.03) Relationship problemsiloss Any relationship problemfloss ?.948 (42.4) 3,?26 (39.5) 4,222 (45.1) ?0.01 1.3 1.3 (1 Intimate partner problem 5,098 2,2?0 (24.1) 2,328 (30.2) <0.01 1.4 1.4 (1 Perpetrator of interpersonal violence in past month 414 (2.2) 131 (1.4) 283 (3.0) 40.01 2.2 2.0 (1.6-2.4) Victim of interpersonal violence in past month 84 (0.4) 53 (0.6) 31 (0.3) <0.05 0.6 (0 4?0.9) 0.8 Family relationship problem 1,6?1 (3.9) 3?3 (9.3) 793 (8.5) 0.9 1.0 Other relationship problem (nonintimate) 403 (2.1) 202 (2.1) 201 (2.1) 1.0 1.1 Argument or con?ict (not specified) 2,914 (15.5) 1,2?3 (13.6) 1,636 (1 <0.01 1.3 1.4 (1 Death of a loved one (any) 1,49? (8.0) 326 (3.3) 6?1 <0.01 0.3 0.9 (0.3-0.9) Nonsuicide death 1,181 (6.3) 64? (6.9) 534 <0.01 0.3 0.9 Suicide of family or friend 379 (2.0) 21? (2.3) 162 <0.01 0.8 Other life stressors Any life stressor 9,?43 (51.9) 4,6?5 5,068 (54.2) <0.01 1 2 1 1 (1 Recent criminal legal problem 1,588 (8.5) 586 (6.2) 1,002 <0.01 1 3 (1.6-2.0) 1 (1.5-1.9) Other legal problem ?48 (4.0) 3?8 (4.0) 370 (4.0) 1 0 (0.3?1 .1) 1.0 (0.9-1.2) Physical health problem 4,1?9 (22.3) 2,012 (21.4) 2,16? (23.2) <0.01 1 1 1.0 (1 JobiFinancial problem?? 2941 (16.2) 1530 (16.3) 1411 (15.6) ??0.05 0 9 0.9 Eviction or loss of home ?22 (3.3) 31? (3.4) 405 (4.3) <0.01 1.3 1.4 (1 School problem?" 162 (19.9) ?0 (1 92 (21.9) 1.3 1.3 (0.9-1.9) Recent release from an 1,412 941 (10.2) 4?1 (5.1) <0.01 0.5 0.5 (0.4-0.5) JailiPrisoniDetention facility 203 (14.4) 32 121 ??0.01 3.6 4.5 Hospital 51? (36.6) 311 (33.0) 206 <0.01 1.6 1.3 (1 hospitaliinstitution 469 (33.2) 439 30 (6.4) <0.01 0.1 0.1 (0.1411) Other (includes alcohoIISA treatment facilities) 223 (15.3) 109 (11.6) 114 (24.2) <0.01 2.4 2.5 (1.3-3.3) Recent or Impending Crisis Crisis within past or upcoming 2 weekst?r?r 5,525 (29.4) 2,444 (26.0) 3,031 (32.9) ??0.01 1.4 1.4 (1 Intimate partner problem crisis 1968 (35.6) 354 (34.9) 1114 (36.2) 1.1 1.1 Physical health problem crisis "(39 (13.4) 315 (12.9) 424 (13.8) 1.1 1.0 (0.3-1.2) Criminal legal problem crisis 621 (11.2) 203 (8.3) 418 (13.6) <0.01 1.6 (1 Family relationship problem crisis 430 (7.3) 212 213 <0.05 0.3 0.9 Job problem crisis 354 (6.4) 191 163 (5.3) <0.01 0.7 Suicide eventfhistory Left a note 6,468 (34.5) 3,132 (33.8) 3,286 (35.1) 1.1 (1.0?1 .1) 1.2 (1 .1?12) Disclosed suicide intent 4.405 (23.5) 2,306 (24.5) 2,099 (22.4) <0.01 0.9 0.9 (0.3-0.9) History of ideation 5,990 (31.9) 3,833 (40.3) 2,152 (23.0) <0.01 0.4 0.4 History of attempts 3,?32 (19.9) Q94) 962 (10.3) <0.01 01303?03) 0.3 Abbreviations: ADDFADHD attention de?cit disorderi'attention deficit hyperactivity disorder; OR odds ratio; PTSD posttraumatic stress disorder; SA substance abuse. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health problem relative to those with known mental health problem. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health problem was the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. it Not significant. ?t Denominator is decedents aged :18 years. Tm Denominator is decedents aged 10?13 years. Denominator of institution subgroup is deoedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonlv occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age-adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler. Scott R. chler. Kcming Yuan. Kristin M. Holland. Asha Z. [trey-Stephenson. Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National 1Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases P3093. Rates increased significantly among males and females in 34 and 43 states, respectively. Fifty-four percent of decedents in 27 states in 2015 did not have a known mental health condition. Among decedents with available information, several circumstances were significantly more likely among those without known mental health conditions than among those with mental health conditions, including relationship problemsiloss [45.1% versus life stressors (54.2% versus and crises [32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family members in the aftermath of a suicide. introduction In 2016, nearly 45,000 suicides (15.6f100,000 population [age-adjusted]) occurred in the United States among persons aged 210 years (I). From 1999 to 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide rates have also increased among persons in all age groups <75 years. with the highest?l argest percent increase (45?1? from [3.2 per Hit] 000 persons ll999l to 19.2 per 100. 000 [2016]} andthe greatest numbel of suicides {n=232. I08 between I999 and 2016] occurring among adults aecd 45 64 veazs. Suicide [5 the 10th leading cause of death and 1s one of just three leading causes that are increasing 4). In addition rates of emergency department visits for nonfatal self- harm. a main risk factor for suicide, increased 42% from 2001 to 2016 (I 1. Together, suicides and self?harm injuries cost the nation approximately $69 billion in direct medical and work loss costs (I l. The National Strategy for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyfrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather. is determined by multiple factors. Despite SSP guidance. suicide prevention largely focuses on mental health conditions alone by identifying suicidal persons. providing treatment for mental health conditions. and preventing rcattempts (6). In addition to mental health conditions and prior suicide attempts. other contributing circumstances include social and economic problems. access to lethal means substances. firearms} among persons at risk. and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal. established by the National Action Alliance of Suicide Prevention and the American Foundation for Suicide Prevention. of reducing the annual suicide rate 20% by 2025 To assist states in achieving this goal. CDC analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 3:10 years?only, as determining suicidal intent in younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (International Classification of Diseases, Tenth Revision, underlying-eause-of death codes Y87.0. U03). Age-specific population estimates were obtained from US. Census Bureaur?National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999-2016 (1999?2001; 2002 ?2004; 2005-2007'; 2008-2010; 2011-2013: and 2014? 2016). Rate estimates were age-adjusted to the US. 2000 standard population and expressed per 100.000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates. employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes. Characteristics of persons aged 31.0 years who died by suicide. with and without known mental health conditions. and the circumstances surrounding the suicides were compared in the 2? states* with complete data participating in National Violent Death Reporting System in 2015. de?nes mental health conditions as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders. Fifth Edition (9). with the exception of alcohol and other substance use disorders. which are captured separately in aggregates data from three primary data sources: death certi?cates. coroner/medical examiner reports {including toxicology}. and law enforcement reports. A range of circumstances (greladonship problems. life stressors. and recent or impending crises} hasve been identified as potential risk factors for suicide. Circumstances captured by are those identified by next of kin as having actively contributed to a person?s suicide. Decedents could have experienced multiple circumstances. Decedents with and without known mental health conditions were compared using Chi- square tests. Logistic regression analyses estimated adjusted odds ratios (aORs) with 95% confidence intervals (CIs). controlling for age group. sex. and racei?ethnicity. Results The most recent overall suicide rates (representing 2014?2016) varied fourfold. from 6.9 (DC) to 29.2 {Montana} per 100.000 persons per year (Supplementary Table; Across the study period. rates increased in all states except Nevada (where the rate was consistently high throughout the study period). with absolute increases ranging from 0.8 per 100.000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota). with increases observed in 25 states (Supplementary Table; (Figure). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and Females {43 states}, as well as for the United States overall (Supplementary Table; Nationally. the model-estimated average annual percentage change for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated signi?cant average annual percentage change increases for males and females (Supplementary Table; Suicide decedents without known mental health conditions (11,039) were compared with those with known mental health conditions (9,407) in 27 states. Whereas all decedents were predominately male (Table l) and non-Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 95% and belong to a racialtethnic minority (OR range: Suicide decedents without known mental health conditions also had signi?cantly higher odds of perpetrating homicide followed by -suicide 2.9, 95% CI: Among adult decedents aged 218 years, 20.1% of those without known mental health conditions and 15.3% of those with mental health conditions had ever served in the military or were serving at the time ofdeath. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health conditions. decedents without known mental health conditions, relative to those with known mental health conditions, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained significant in the adjusted models. Toxicology testing was less likely to be performed for decedents without known mental health conditions. Among those with toxicology results, decedents without known mental health conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, Cl Information on circumstances surrounding suicide were available for all decedents with mental health conditions (n 9,407) and approximately 85% ofthose without known mental health conditions (n 9,357) in 27 states (Table 2). Persons without known mental health conditions were less likely to have any substance use disorders 0.7, 95% CI 0.7?0.8) than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment just over half were in treatment at the time of death. Deccdents without known mental health conditions had a signi?cantly higher likelihood of any relationship problem/loss than did those with known mental health conditions speci?cally intimate partner problems (30.2% versus (17.5% versus and recendy perpetrating interpersonal violence in the past month versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (54.2% versus 49.7%} such as criminalflegal problems (10.7% versus or evictiontloss ofhome versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis {a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Physical health problems and job?e?a?nancial problems were commonly contributing stressors among both persons without mental health conditions (23.2% and 15.6%, respectively) and those with mental health conditions (21.4% and 16.8%, respectively). Similarly, among all persons with recent crises. intimate partner problems were the most common types and did not differ by group. Decedents without known mental health conditions had signi?cantly lower odds of recent release from any institution 0.5, 95% CI Among those recently released, decedents without known mental health conditions were signi?cantly more likely than decedents with mental health conditions to have been released from a correctional facility (25.7% versus hospital (43.7% versus or Among with known mental health conditions who were recently released from an institution. 4am were released from facilities. other facility keg. treatment] (24. 2% versus Im?l? Should this be two sets] at parentheses or just one? without known mental health conditions were signi?cantly less likely to have a history of suicidal ideation [23. ti" it or prior suicide attempts It] compared with dtose with known mental health conditions {40. 8% and 29 40/ 0 respectively). Suicide intent was disclosed persons without and with known mental health conditions, respectively. Conclusions and Comments During ?99?20%. suicide rates increased signi?cantly in 44 states, and 25 states experienced Increases Rates increased signi?cantly among males in 34 states, and females in 43 states. Additional research into the speci?c cattses of these trends is needed. Data from the 2? states participating in provide important insight into circutnstanees surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single lbctor however, suicide prevention is often oriented toward mental health conditions alone vis?t?I?vis identi?cation of suicidal persons. treatment of mental health conditions. and prevention of This study found that approximately half of suicide decedents in did not have a known mental health condition. indicating that additional focus on nonmental health factors further upstream could provide important information for a public health approach till). Those without a known mental health condition suffered more -from relationship problems and other life stressors such as criminaly?legal matters. evictiontloss of home. and recent or impending crises. Similarly, persons with mental health conditions also often experienced health factors such as relationship problems and other life stressors such as jobt?nancial or physical health problems. These ?ndings point to the need to thl-p- pennants- manage mental? health-eoodit-i?ons? and; to prevent the circumstancos associated with their onset oi" mental health conditions in the first place: as?vrell?l support persons with known mental health conditions to decrease their risk for poor outcomes [1 Two thirds of suicide decedents with mental health conditions had a history of treatment for any?mental health ggLJ-{or substance with approximately halt?in treatment when they died. This ?nding suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities. such as doctor?patient collaborative care models and proven cognitive?behavioral therapies. In addition. increased access to behavioral health providers in areas is needed. as is expansion of health care systems that integrate physical and behavioral health. with a priority on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide Prevention strategies include strengthening economic supports (cg housing stabilization policies. household ?nancial support): teaching coping and problem- solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense ofbelonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk (cg, military veterans. persons with physicaltmental health conditions) Other strategies include creating protective environments (cg. reducing access to lethal means among persons at risk of suicide, creating organizational and workplace policies to promote help?seeking, easing transitions into and out of work for persons with mental health conditions and other life challenges]. strengthening access and delivery of care- For- persons- air risk, supporting family and friends after a suicide. and assuring the media t?ollow sat?e reporting recommendations Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention to}. The ?ndings in this report are subject to at least three limitations. First, in the state-level analysis, rankings for four states {Maryland, Massachusetts, Rhode Island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative: the 27 states included represent 49.6% of the population Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of-kin o?en identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown. and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policy, Programs, and Practices (12) to better understand suicide in their populations, prioritize evidencc~based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard, Margaret Warner, Division of Vital Statistics, National Center for Health Statistics, CDC. Con?ict of Interest No con?icts ofinterest were reported. 1Division of'v'iolence Prevention. National Center for Injury Prevention and Control. CDC r'Division ofAnalysis. Rasearch. and Practice Integration. National Center for Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone. 717041383942. References l. CDC. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. AZ, Crosby AE, Jack SPD, I-lailcyesus T, Suicide trends among and within urbanization levels by sex, racei?ethnieity, age group, and mechanism of death?United States, 2001?2015. MMWR Surveill Summ 2017;66{No, 83-18). 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. NCHS data brief no 241. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. 4. Kochanek K, Murphy 5, Xu J, Arias E. Mortality in the United States, 2016. NCHS data brief no 293. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics: 2017. 5. Of?ce of the Surgeon General: National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action. Washington, DC: US Department: of Health and Human Services, Of?ce of the Surgeon General; 2012. rcport.pdf 6. Zalsman G, Hawton K, Wasserman D, et a1. Suicide prevention strategies revisited: 10-year systematic review. Lancet 7. Torguson K, O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington, DC: American Foundation for Suicide Prevention; 2017. 8. Crepeau-l?Iobson F. The autopsy,r and determination ot?child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders (DSM- 5). Washington, DC: American Association; 2013. 10. Caine ED, Reed J, Hindrnan J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017. Epub December 20, 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM, Holland KM, Bartholovv BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package of policy, programs, and practice. Atlanta, GA: US Department of Health and Human Services, 2017'. l3. Milner A, Sveticic J, De Lee D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona, Colorado. Connecticut, Georgia. Hawaii, Kansas, Kentucky. Maine, Maryland. Massachusetts, Michigan, Minnesota. New Hampshire, New Jersey. New Mexico. New York. North Carolina. Ohio, Oklahoma, Oregon, Rhode island. South Carolina, Utah, Vermont, Virginia, and Wisconsin. Summary What is already known about this topic? In 2015, hearty 45,000 persons died by suicide in the United States. Mental health conditions can contribute to suicide. What is added by this report? During 1999?2016, suicide rates increased in nearly,r every state, including >30% increases in 25 states. 2015 data from 2? states indicate 54% of suicide decedents were not known to have mental health conditions. Other contributors included relationship, substance use, health, and job or ?nancial problems, among others. What are the implications for public health practice? A comprehensive approach using proven prevention strategies. such as those in (300's Technical Package for Suicide Prevention. can help reach the national goal of reducing the annual suicide rate 20% by 2025. TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged :19 I(rears with and without known mental health conditions - National Violent Death Reporting System. 21' states.* 2915 Known mental No known mental Total health condition1 health condition Adjusted Characteristic (N 29.448) (n 9.49?) (n =11.939) P-value (95% CI) (95% Sex Male 15.?92 (78.8) 8.489 (88.8) 9.233 (83.8) {9.91 2.3 NA Female 4.?44 (23.2) 2.938 (31.2) 1.898 (18.4) {9.91 9.4 (9.445) NA Ase ?rst" 19?24 2.894 1.211 (12.9) 1.593 (14.4) {9.91 1.1 (1.1~1 .2) NA 25?44 8.458 (31.8) 3.938 (32.3) 3.429 (31.9) {9.95 9.9 NA 45?84 3.829 (49.8) 3.898 (35.3) {9.91 9.8 NA 285 3.488 1.349 (14.2) 2.128 (19.3) {9.91 1.4 (1.3-1.5) NA White. non~Hispanic 1?.192 (83.8) 8.185 (88.8) 8.93? (81.9) {9.91 9.8 NA Eilacic. non-Hispanic 1.225413 411 (3.4) 8137.4) {9.91 NA American Indiaanlaska Native. non?Hispanic 3?8 (1.8) 112 (1.2) 288 (2.4) {9.91 2.9 NA Asian. non-Hispanic: 5?8 (2.8) 235 (2.5) 341 (3.1) {9.95 1.2 (1.1-1.5) NA Hispanic 1.998 (5.4) 483 (4.9) 833 {9.95 1.2 (1.9-1.3) NA Other 88 (9.3) 21 (9.2) 45 (9.4) {9.95 1.8 NA Extended demographics Ever served in militant?r 3.429 1.354 (15.3) 2.9?5 (29.1) {9.91 1.4 1.1 (1.9?1 .1) Homeless 249 (1.2) 194 (1.1) 138 (1.3) 1.1 (9.9-1.5) 12 Incident Type Single suicide 29.98%982 9.318 (99.1) 19.?45 {9.91 9.3 9.4 Homicide followed by suicide 319 (1.8) 84 255 (2.3) {9.91 3.5 2.9 Multiple suicides 84 (9.3) 25 (9.3) 39 (9.4) NS 1.3 1.8 Method Firearm 9.999 (48.5) 3.821 (49.8) 8.988 (55.3) {9.91 1.8 5.99? (28.9) 2.949 (31.3) 2.98? (28.9) {9.91 9.8 9.8 (9.?498) Poisoning 3.993 1.142 Q94) {9.91 9.8 Substance class causing death? Odie-r (9.9-. over-me-counter) 1.921 (34.9) 888 (35.8) 355 (31.1) {9.91 9.8 9.9 Opioids 944 (31.4) 898 338 (29.4) NS 9.9 9.9 (9.8?1 .1) Antidepressants 899 (28.8) 844 (34.8) 158 {9.91 9.3 9.3 Benzodiazepines 824 (29.8) 488 (25.1) {9.91 9.5 9.5 219 (7.3) 195 (19.5) 24 (2.1) {9.91 9.2 (9.1-9.3) 9.2 (9.1-9.3) Other 1.595 (7.8) 789 (8.3) 815 {9.95 9.9 9.9 Toxicology Results Anltoxicology testing 13.31? (85.1 8.858 (T98) 8.859 (89.3) {9.91 9.8 Positive for 21 substance111 9.913 (74.4) 5.192 4.?21 (79.9) {9.91 9.8 Substance detected?? Alcohol Tested 19.959 (53.8) 5.499 5.541 (59.2) {9.91 9.7 9.8 Positive 4.442 (49.8) 2.115 (39.1) 2.32? (42.9) {9.91 1 .1 1.2 Opioids Tested 8.554 4.258 (45.3)l 4.298 (38.9)l {9.91 9.8 9.8 (9.8-99) Positive 2,2?9 1,238 1,041 {24.2? <0.01 0.8 0.9 Benzodiazepines Tested 8,124 4,228 {44.9) 3,898 {35.3) {0.01 031108?03) {01?08) Positive 2,454 {30.3} 1.539 {38.8) 825 {21.2) <0.01 0.4 0.5 Cocaine Tested {39.0) 3,886 (41.1) 4,112 {312) <0.01 0.9 0.9 Positive 499 215 283 <0.05 1.2 1.2 Amphetamines Tested 1515 {312} 3.698 {39.3) 3,919 {35.5) <0.01 0.9 (0.8-0.9) 0.9 Positive 738 378 (10.2) 380 (9.2) NS 0.9 1.0 (0.8?1 .1) Marijuana Tested 8,559 {32.1} 312713382) 3,442 {31.2) <0.01 0.9 0.9 Positive 1,4?1 {22.4) T10 T81 {22.1) NS 1.0 0.9 (0.8-1.0) Antidepressants Tested 5,425 {28.5) 3,103 (33.0) 2,322 {21.0) {0.01 0.5 0.8 Positive 2,214 {40.8) 1,?35 {55.9) 479 {20.8) <0.01 0.2 0.2 Abbreviation: CI con?dence interval; NA not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health condition in ooronerimedical examiner or law enforcement reports. ?93 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. YT Denominator is decedents aged 218 years with reported military service status. ?t Denominator is decedents who died by poisoning, including overdose. rm Denominator is decedents with any toxicology testing. Denominator for each positive group is the number tested for the substance in that group. TABLE 2. Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental health No known mental health conditiont condition P- Adjusted (95% Characteristic Total no. no. value OR5 {95% Cl} Cl) Suicide with known circumstances 18,754 (91.8) 9,407 {100) 9,35? {84.8) ?0.01 NA NA Mental health Any Current Diagnosed Mental Health Condition" T075 NA NA NA NA Anxiety disorder 1.5T9 {18.8) NA NA NA NA Bipolar disorder 1,431 {15.2) NA NA NA NA Schizophrenia 509 NA NA NA NA PTSD 424 NA NA NA NA ADDIADHD 225 NA NA NA NA Ed Substance Alcohol 1 1 1 403 15.0 . 0.1" 3?0.8 1.?68 3.3 31 14. El 7-413 Current mental healtl?li'substanoe abuse treatment . 5 0.01 .0 for 6? 394 . 0.0? 0.02 45.1 . . . 2-1.4 NEE in month nonihtimate Death of a loved one death fa Other stressars life stressor Other Dr 1053 home Recent crisis Criminal relations ave Disclosed suicide intent 4 405 4.5 (0.01 . 1.0 Histo of ideation 1.9 3 40.3 . of 19 .4 40.01 . Abbreviations: attention deficit disorderiattention disorder; A not stress SA substance abuse. Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South Carolina. Utah. Vomtont. Virginia. and Wisconsin. Teresalthihk allaf the diagnoses need to he indented to better indicate that Torrent depressed mood? is not a diagnosis {but it still falls under mental health more broadly}. Decedent had been identified as having a current diagnosis of mental health condition in coroneri'medical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. it Denominator is decedents aged 218 years. Denominator is decedents aged 10?1 3 years. Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks, Crises depicted here represent the most commonly occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age?adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. Thomas R. Simon Katherine A. Fowler, Scott R. Kegler. PhDiz Kcming Yuan. Kristin M. Holland. Asha Z. Hey-Stephenson. Alex E. Crosby, MIDI1 Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are just one factor contributing to suicide. Examining state-level trends in suicide and the multiple circumstances contributing to it can infomi comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persoms aged 210 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data from the National 1'v?ital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states. with 25 states experiencing increases 80%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. Fifty-four percent of decedents in 27 states in 2015 did not have a known mental health condition. Among decedents with available information. several circumstances were signi?cantly more likely among those without known mental health conditions than among those with mental health condition, including relationship problemsfloss [45.1% versus life stressors (54.2% versus and crises [32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased significantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family members in the aftermath of a suicide. Introduction In 2016, nearly 45,000 suicides (15.6f100,000 population [age-adjusted]) occurred in the United States among persons aged 210 years (I). From 1999 to 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide rates have also increased among persons in all age groups <75 years, with the highest percent increases among those aged 10-14 years (76% increase from 1.2 to 2.1 per 100,000 in 1999 and 2016, respectively) and those aged 45-64 years a(45% increase from 13.2 to 19.2 per 100,000 in 1999 and 2016, respectively) (1.3). Suicide is the 10th leading cause of death and is one of just three leading causes that are increasing In addition. rates of emergency department visits for nonfatal self?harm. a main risk factor for suicide, increased 42% from 2001 to 2016 (3). Together. suicides and self-harm injuries cost the nation approximately $69 billion in direct medical and work less costs (I). The National Strategy for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with. efforts spanning multiple levels individual, familytrelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factor, but rather, is determined by multiple factors. Despite NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health conditions, and preventing rcattempts In addition to mental health conditions and prior suicide attempts, other contributing circumstances include social and economic problems, access to lethal means substances, ?rearms) among persons at risk, and poor coping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the national goal, established by the National Action Alliance of Suicide Prevention and the American Foundation for Suicide Prevention, of reducing the annual suicide rate 20% by 2025 To assist states in achieving this goal, CDC. analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide: this report presents options for multilevel comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 210 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certi?cate records (International Classification af' Diseases, Tenth Revisian. underlying-causevof death codes Y87.0, U03). population estimates were obtained from U.S. Census Bureaquational Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive 3-year aggregate periods spanning 1999~-2016 0999-2001; 2002u2004; 2005-4007; 2008?2010; 2011?2013; and 2014- 2016). Rate estimates were age-adjusted to the US. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes. Characteristics of persons aged :10 years who died by suicide, with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 2? states* with complete data participating in CDC ?s National Violent Death Reporting System in 2015. de?nes mental health conditions as disorders and listed in the Diagnostic and Statistical flair-ritualr of Mental Disorders, Fain Edition with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certi?cates, coroner/ med ical examiner reports including toxicology), and law enforcement reports. A range of circumstances (relationship problems, life stressors, and recent or impending crises) have been identi?ed as potential risk factors for suicide. Circumstances captured by are those identified by next of kin as having actively contributed to a person?s suicide. Decedent's could have experienced multiple circumstances. Decedents with and without known mental health conditions were compared using Chi- square tests. Logistic regression analyses estimated adjusted odds ratios (?10st with 95% confidence intervals (C Is), controlling for age group, sex, and racci?ethnicity. Results The most recent overall suicide rates (representing 2014?2016) varied fourfold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from 0.8 per 100,000 (Delaware) to 8.1 (Wyoming). Percentage increases in rates ranged from 5.9% (Delaware) to 57.6% (North Dakota), with increases observed in 25 states (Supplementary Table; (Figure). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females (43 states}, as well as for the United States overall (Supplementary Table; stacks.cdc. gow?viewfcde/ 5 3785). Nationally, the model-estimated average annual percentage change for the overall suicide rate was an increase of By sex, estimated national rate trends further indicated signi?cant average annual percentage change increases for males and females (Supplementary Table; Suicide decedents without known mental health conditions (11,039) were compared with those with known mental health conditions (9,407) in 27 states. Whereas all decedents were predominately male (Table 1) and non-Hispanic white those without known mental health conditions, relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 95% (31:22?25) and belong to a racialfethnie minority (OR range: Suicide decedents without known mental health conditions also had signi?cantly higher odds of perpetrating homicide-suicide 2.9, 95% C1 2 Among adult decedents aged 218 years, 20.1% of those without known mental health conditions and 15.3% of those with mental health conditions had ever served in the U.S. military or were serving at the time of death. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health conditions, decedents without known mental health conditions, relative to those with known mental health conditions, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained significant in the adjusted models. Toxicology testing was less likely to be performed for decedents without known mental health conditions. Among those with toxicology results, decedents without known mental health conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, C1 Information on circumstances surrounding suicide were available for all decedents with mental health conditions (n 9,407) and approximately 85% of those without known mental health conditions (11 9,357) in 27 states (Table 2). Persons without known mental health conditions were less likely to have any substance use disorders 0.7, 95% CI 0.7?0.8) than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment just over half were in treatment at the time of death. Deccdents without known mental health conditions had a signi?cantly higher likelihood of any relationship than did those with known mental health conditions speci?cally intimate partner problems (30.2% versus argumentsfeonflicts (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (54.2% versus 49.7%) such as criminal/legal problems (10.7% versus or eviction/loss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Physical health problems and job or ?nancial problems were commonly contributing stressors among both persons without mental health conditions (23.2% and 15.6%, respectively) and those with mental health conditions (21.4% and 16.8%, respectively). Similarly, among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Decedents without known mental health conditions had significantly lower odds of recent release from any institution 0.5, 95% C1 Among those recently released, decedents without known mental health conditions were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcoholfsubstance treatment) (24.2% versus than were those recently released with a known mental health condition. Among dccedents with known mental health conditions who were recently released from an institution, 46.7% were released from facilities. Decedents without known mental health conditions were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%. respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health conditions. respectively. Conclusions and Comments During 1999?2016. suicide rates increased signi?cantly in 44 states, and 25 states experienced increases 3,30%. Rates increased signi?cantly among males in 34 states. and females in 43 states. Additional research into the specific causes of these trends is needed. Data from the 27 states participating in provide important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however. suicide prevention is often oriented toward identification of suicidal persons, treatment of mental health conditions. and prevention of reattempts. This study found that approximately half of suicide decedents in did not have a known mental health condition. indicating that additional focus on health factors further upstream could provide important information for a public health approach Those without a known mental health condition suffered more from relationship problems and other life stressors such as criminalflegal matters, evictioni?loss of home. and recent or impending crises. Similarly. persons with mental health conditions often experienced relationship problems and other life stressors such asjobflinaneial or physical health problems. These ?ndings point to the need to help persons manage mental health conditions and to prevent the circumstances associated with their onset. as well support persons with known mental health conditions to decrease their risk for poor outcomes (11). Two thirds of suicide decedents with mental health conditions had a history of treatment for any mental health or substance use or both. with approximately half in treatment when they died. This ?nding suggests the need for additional safety supports. including broader implementation of affordable and effective treatment modalities. such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. In addition. inereased access to behavioral health providers in underserved areas is needed. as is expansion of health care systems that integrate physical and behavioral health. with a priority on suicide prevention and patient safety. especially through care transitions U2). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies. household financial support); teaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems. especially early in life; promoting social connectedness to increase a sense of belonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk military veterans. persons with physicalfmental health conditions} Other strategies include creating protective environments reducing access to lethal means among persons at risk of suicide. creating organizational and workplace policies to promote helpasceking. easing transitions into and out of work for persons with mental health conditions and other life challenges). strengthening access and delivery of care for persons at risk. supporting family and friends after a suicide. and assuring the media follow safe reporting recommendations (1.3). Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention The ?ndings in this report are subject to at least three limitations. First. in the state-level analysis. rankings for four states (Maryland. Massachusetts. Rhodc Island. and Utah) might have been affected by large proportions of injury deaths of undetermined intent {potentially biasing reported suicide rates downward) or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second. is not yet nationally representative; the 27 states included represent 49.6% of the population Finally. abstractors of data are limited to information contained in investigative repoits. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies that include more in- depth interviews with next-of-kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (M). It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown. and hence underrepoited by key informants. Nonetheless. the high prevalence of diverse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package of Policy. Programs. and Practices (i2) to better understand suicide in their populations. prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard. Margaret Warner. Division of Vital Statistics. National Center for Health Statistics, CDC. Con?ict of Interest No con?icts of interest were reported. 'Division of?v?iolence Prevention. National Center For Injury Prevention and lControl. EDivision ofAnalys-is. Research. and Practice Integration. National Center for Injury Prevention and Control. CDC. Corresponding author: Deborah M. Stone. 770-4883 942. References 1. CDC. Web-based injury Statistics Query and Reporting System Atlanta. GA: National Center for Injury Prevention and Control. Retrieved March 15. 2018. In. 2. Ivey-Stephenson AZ. Crosby AE. Jack SPD, Haileyesus T. Kresnow-Sedacca MI. Suicide trends among and within urbanization levels by sex. race/ethnicity. age group. and mechanism of death?United States. 2001?2015. MMWR Surveill Summ 2017:66tNo. SS-IS). 3. Curtin SC. Warner M. Hedegaard H. Increase in suicide in the United States. 1999?2014. NCHS data brief no 241. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics: 2016. .pdf 4. Koehanek K. Murphy S. Xu J. Arias E. Mortality in the United States. 2016. NCHS data brief no 293. Hyattsville. MD: US Department of Health and Human Services. CDC. National Center for Health Statistics; 2017'. 5. Of?ce ot'the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: goals and objectives for action. Washington. DC: US Department ot?Health and Human Services. Of?ce of the Surgeon General; 2012. 6. Zalsrnan G. Hawton K. Wasserman D. et Suicide prevention strategies revisited: 10-year systematic review. Lancet 2016;3z646?59. 7. Torguson K, O'Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington, DC: American Foundation for Suicide Prevention; 2017. 8. Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders (DSM- S). Washington, DC: American Association; 2013. 10. Caine ED, Reed J, Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017. Epub December 20, 2017. 1. World Health Organization. Risks to mental health: an overview of vulnerabilities and risk factors. Geneva, Switzerland: World Health Organization; 2012. 12. Stone DM, Holland KM, Bartholow Crosby AE, Davis SP, Wilkins N. Preventing suicide: a technical package of policy, programs, and practice. Atlanta, GA: US Department of Health and Human Services, 2017. l3. Milner A, Sveticie J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Alaska. Arizona, Colorado. Connecticut. Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan, Minnesota. New Hampshire. New Jersey, New Mexico. New York, North Carolina, Ohio, lDidahoma, Oregon, Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Summaryr 1titl'hat is already known about this topic? In 2015, neariy 45,000 persons died by suicide in the United States. Mental health conditions can contribute to suicide. What is added by this report? During 1999?2015, suicide rates increased in nearly every state, including >30% increases in 25 states. 2015 data from 2? states indicate 54% of suicide decedents were not known to have mental health conditions. Other contributors included relationship, substance use, health, and job or ?nancial problems, among others. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in (300?s Technical Package for Suicide Prevention, can help reach the national goal of reducing the annual suicide rate 20% by 2025. TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged :10 years with and without known mental health conditions - National Death Reporting System, 2? states) 2015 Known mental No known mental Total health condition1 health condition Adjusted Characteristic (N 20,446) (n 9,40?) {n 11,039) F-value 0R5 (95% Cl) (95% Cl} Sex Male 15,?02 6,469 (63.3) 9,233 (33.6) {0.01 2.3 NA Female 4,744 (23.2} 2,933 (31.2) 1,306 (16.4} {0.01 0.4 NA Ase oral? 10?24 2,304 {13.7} 1,211 (12.9) 1.593 (14.4} {0.01 1.1 NA 25?44 6,456 (31.6} 3.036 (32.3) 3,420 (31.0) {0.05 0.9 NA 45?64 3,320 (40.6) 3,393 (35.3) {0.01 0.3 NA 265 3.463 1,340 (14.2) 2,123 (19.3} {0.01 1.4 (1.3-1.5) NA White, non-Hispanic 1?,102 (33.6} 3.165 (36.3) 3,93? (31.0} {0.01 0.6 NA Black, non-Hispanic 1,22?60} 411 31? {0.01 NA American IndianiAlaska Native, non?Hispanic 3?8 112 (1.2) 266 (2.4) {0.01 2.0 NA Asian. non-Hispanic 576 235 (2.5) 341 {0.05 1.2 (1.1-1.5) NA Hispanic 1.096 463 633 {0.05 1.2 NA Other 66 21 (0.2) 45 {0.05 1.3 NA Extended demograghics Ever served in military,'11 3,429 1,354 (15.3) 2,0?5 (20.1) {0.01 1.4 1 1 (1.0?1 1) Homeless 240 104 (1.1) 136 1.1 12 (0.9-1 5} incident Type Single suicide 20,063 (93.2} 9,313 (99.1) 10,?45 {0.01 0.3 0.4 Homicide followed by suicide 319 64 255 {0.01 3.5 2.9 Multiple suicides 64 25 (0.3) 39 (0.4) NS 1.3 1.6 Method Firearm 9.909 (43.5} 3.321 (40.6) 6.033 (55.3} {0.01 1.3 1.6 5,90? (23.9} 2,940 (31.3) 2,96? (26.9) {0.01 0.3 0.3 Poisoning 3,003 1,361 (19.3) 1,142 (10.4) {0.01 0.5 0.6 Substance class causing death?? Other over-the-counter} 1,021 (34.0} 666 (35.3} 355 (31.1} {0.01 0.3 0.9 Opioids 944 (31.4} 603 336 (29.4} NS 0.9 0.9 (0.3?1 .1) Antidepressants 300 (26.6) 644 (34.6) 156 {0.01 0.3 0.3 Benzodiazepines 624 (20.3} 463 (25.1) 156 {0.01 0.5 0.5 219 195 (10.5) 24 {0.01 0.2 0.2 Other 1,595 730 315 {0.05 0.9 0.9 Toxicology Results Any.r toxicology testing 6,653 (T03) 6,659 (60.3) {0.01 0.6 Positive for 21 substance?? 9,913 5,192 4,?21 (1'09) {0.01 0.3 Substance Alcohol Tested 10,950 (53.6} 5.409 5,541 (50.2} {0.01 0.3 Positive 4,442 (40.6} 2,115 (39.1) 2,32? (42.0} {0.01 1.1 1.2 Opioids Tested I 3,554 4.253 4,296 {0.01 0.3 0.3 Positive 2,2?9 1,238 1,041 {24.2? <0.01 0.8 0.9 Benzodiazepines Tested 8,124 4,228 {44.9) 3,898 {35.3) {0.01 031108?03) {01?08) Positive 2,454 {30.3} 1.539 {38.8) 825 {21.2) <0.01 0.4 0.5 Cocaine Tested {39.0) 3,886 (41.1) 4,112 {312) <0.01 0.9 0.9 Positive 499 215 283 <0.05 1.2 1.2 Amphetamines Tested 1515 {312} 3.698 {39.3) 3,919 {35.5) <0.01 0.9 (0.8-0.9) 0.9 Positive 738 378 (10.2) 380 (9.2) NS 0.9 1.0 (0.8?1 .1) Marijuana Tested 8,559 {32.1} 312713382) 3,442 {31.2) <0.01 0.9 0.9 Positive 1,4?1 {22.4) T10 T81 {22.1) NS 1.0 0.9 (0.8-1.0) Antidepressants Tested 5,425 {28.5) 3,103 (33.0) 2,322 {21.0) {0.01 0.5 0.8 Positive 2,214 {40.8) 1,?35 {55.9) 479 {20.8) <0.01 0.2 0.2 Abbreviation: CI con?dence interval; NA not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health condition in ooronerimedical examiner or law enforcement reports. ?93 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. YT Denominator is decedents aged 218 years with reported military service status. ?t Denominator is decedents who died by poisoning, including overdose. rm Denominator is decedents with any toxicology testing. Denominator for each positive group is the number tested for the substance in that group. TABLE 2. Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental health No known mental health conditiont condition P- Adjusted (95% Characteristic Total no. no. value OR5 {95% Cl} Cl) Suicide with known circumstances 18,754 (91.8) 9,407 {100) 9,35? {84.8) ?0.01 NA NA Mental health Any Current Diagnosed Mental Health Condition" T075 NA NA NA NA Anxiety disorder 1.5T9 {18.8) NA NA NA NA Bipolar disorder 1,431 {15.2) NA NA NA NA Schizophrenia 509 NA NA NA NA PTSD 424 NA NA NA NA ADDIADHD 225 NA NA NA NA Not specified ?80 NA NA NA NA Current depressed mood 3,982 {42.1) 3,0?8 {32.9) ?0.01 Substance problems Any Current substance problem 5,319 {28.3) 2,9?8 {31.8} 2,343 {25.0) ?0.01 Alcohol problem 3,288 1,882 {19.8) 1,408 {15.0) ?0.01 Other substance problem 3.084 {18.4) 1,?88 {18.8) 1,318 {14.1) ?0.01 Treatment Current mental abuse treatment 5,141 {54.0) 84 {0.01 0.01 (0.01?0.01) 0.01 (0.01?0.01) Ever treated for mental healthi'substanoe problem {35.8) 8,323 394 ?0.01 0.02 (0.02-0.02) 0.02 (0.02?0.03) Relationship problems?oss Any relationship problemiloss ?.948 {42.4) 3.?28 {39.8) 4,222 {45.1} ?0.01 1.3 {1 1.3 {1 Intimate partner problem 5,098 2,2?0 {24.1) 2,828 {30.2) ?0.01 1.4 {1 1.4 {1 Perpetrator of interpersonal violence in past month 414 131 283 {0.01 2.2 {1 2.0 Victim of interpersonal violence in past month 84 53 31 ?0.05 0.8 0.8 Family relationship problem 1,8?1 8?3 ?98 NS 0.9 1.0 Other relationship problem {nonintimate} 403 202 201 NS 1.0 1.1 Argument or con?ict {not specified) 2,914 {15.5) 1,2?8 {13.8} 1,838 {1 ?0.01 1.3 {1 1.4 {1 Death of a loved one {any} 1,49? 828 8?1 <0.01 0.8 0.9 Nonsuicide death 1,181 84? 534 ?0.01 0.8 0.9 Suicide of family or friend 3?9 21? 182 ?0.01 0.8 Other life stressors Any life stressor 9,743 {51.9) 4,8?5 5,088 {54.2) ?0.01 1.2 1 1 Recent criminal legal problem 1,588 588 1,002 {0.01 1.8 1 Other legal problem ?48 3?8 3?0 NS 1.0 1.0 Physical health problem 4,1?9 {22.3) 2,012 {21.4) 2,18? {23.2} ?0.01 1.1 {1 1.0 {1 JobiFinancial problemTl 2941 {18.2) 1530 {18.8) 1411 {15.8) ?0.05 0.9 0.9 Eviction or loss of home ?22 31? 405 ?0.01 1.3 {1 1.4 {1 School problem??i 182 {19.9) ?0 92 {21.9) NS 1.3 1.3 Recent release from an institution? 1,412 941 {10.2) 471 ?0.01 0.5 0.5 JailfPrisoniDetention facility 203 {14.4) 82 121 {25.7) ?0.01 3.8 {27?49) 4.5 Hospital 51? {38.8) 311 {33.0) 208 ?0.01 1.8 1.3 {1 hospitaliinstitution 489 (33.2) 439 30 ?0.01 0.1 0.1 {0.1411} Other {includes alcoholiSA treatment facilities) 223 (15.8) 109 {11.8) 114 {24.2) {0.01 2.4 2.5 Crisis within past or upcoming 2 5,525 {29.4) 2,444 {28.0) 3.081 {32.9) ?0.01 1.4 {1 1.4 {1 Intimate partner problem crisis 1988 {35.8) 854 {34.9) 1114 {38.2) NS 1.1 1.1 Physical health problem crisis ?39 {13.4) 315 {12.9} 424 {13.8) NS 1.1 1.0 Criminal legal problem crisis 821 (11.2) 203 418 {13.8) ?0.01 {15-21) 1.8 {1 .3?19) Family relationship problem crisis 430 212 218 ?0.05 0.8 0.9 Job problem crisis 354 191 183 ?0.01 Suicide eventfhistory Left a note 8,488 {34.5) 3.182 {33.8) 3,288 {35.1) NS 1.1 {1 1.2 {1 Disclosed suicide intent 4,405 {23.5) 2,308 {24.5) 2,099 {22.4) ?0.01 0.9 0.9 History of ideation 5.990 {31.9) 3.838 {40.8) 2,152 {23.0) ?0.01 0.4 0.4 History of attempts 3,?32 {19.9) {29.4) 982 {10.3) ?0.01 0.3 0.3 Abbreviations: ADDIADHD attention de?cit disorderiattention deficit hyperactivity disorder; NA not applicable; OR odds ratio; PTSD posttraumatic stress disorder; SA substance abuse. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identified as having a current diagnosis of mental health condition in coroneri'medical examiner or law enforcement reports. ?95 OR re?ects the risk among those without known mental health condition relative to those with known mental health condition. 1T Logistic regression was used to estimate adjusted OR with 95% after controlling for age, sex, race. and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclusive. Therefore, sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health conditions. it Denominator is decedents aged 218 years. Denominator is decedents aged 10?1 3 years. Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks, Crises depicted here represent the most commonly occurring categories. FIGURE. Percentage change in annual suicide rate,* by state United States, from 1999?2001 to The figure above is a map of the United States showing the percentage change in annual suicide rate, by state, from 1999 to 2016. Per 100,000 population, age?adjusted to the 2000 U.S. standard population. Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Pth; Keming Yuan, Kristin M. Holland, Asha Z. Ivey?Stephenson, Alex E. Crosby, Background: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple circumstances contributing to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive three-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known mental health problem. Among decedents with circumstance information available, several circumstances were signi?cantly more likely among those without a known mental health problems than among those with mental health problems, including relationship problemsfloss [45.1% versus life stressors (54.2% versus and recentfimpending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. Introduction Background and Purpose In 2016, nearly 45,000 suicides (15.6! 100,000 population [age?adjustedD occurred in the United States among persons aged :10 years (1). Between 1999 and 2015, suicide rates increased among both sexes, all raciali?ethnic groups, and all urbanization levels Suicide is the 10th leading cause of death and is one of just three leading causes that are increasing Additionally, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased 42% between 2001 and 2016 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The Neiionei iraiegjv for Suicide Prevention (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyirelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factOr, but rather, is determined by multiple factors. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems, and preventing reattempts (6). In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic problems, access to lethal means (cg, substances, ?rearms) among persons at risk, and poor I contributing to suicide risk and action to address them can help reach he nation?s goal of reducing suicide rates by 20% by 2025} To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multi~level comprehensive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 2 [1 years only. as determining suicidal intent in ii younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on ii National Vital Statistics System coded death certi?cate records (Interactions! CittSsifit'at?ion of Diseases It?? Roi-vision. underlying-eause-ofdeath codes YETO. U03). Age-speci?c population estimates were obtained from US. Census Bureaut?National Center for Health Statistics bridged?race population data releases. National and state~level suicide rate estimates were calculated for six consecutive three?year aggregate periods spanning i999?2ill? 1999?2011], 20(12?2004; 2005?91107; 2008?20 IE 20] l? 2ti13; and 2t] l4?2tll Rate estimates were age-adjusted to the U3. year 2000 standard I population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes Characteristics of persons aged 210 years who died by suicide, with and without knot-rm mental health problems. and the circumstances surrounding the suicides were compared in the 27 states 3 Comment IOADSI: Please identify the agency, organization, or other source for coping and problem-solvin skills Expanded awareness of these additional circumstances . 2020, mass, other. Level 1 i Comment Please correct error: a his national goal, e.g. Healthy People should be 2U10. Level 1. Comment I have alreedv corrected. with complete data participating in National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certi?cates, coronen?medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known mental health problems were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% con?dence intervals (CI), controlling for age group, sex, and race! ethnicity. Results The most recent overall suicide rates (representing 2014?2016) varied four-fold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table). Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from +0.8 per 100,000 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (Supplementary Table) (Figure 1). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females (43 states), as well as for the United States overall (Supplementary Table). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated signi?cant increases for males (AAPC +1 . and females (AAPC (Supplementary Table). Suicide decedents without known mental health problems (N 11,039) were compared with those with known mental health problems (N 9,407) in 27 states. Whereas all decedents were predominately male (768%) (Table 1) and non?Hispanic white those without known mental health problems, relative to those with mental health problems, were more likely male (83.6% versus 68.8%; odds ratio (OR) 2.3, 95% CI 2.2?2.5) and racial/ethnic minorities (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI 2.2? 3.8). Among adult decedents aged 3:18 years, 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever served, or were currently serving, in the U.S. military. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health problems, decedents without known mental health problems, relative to those with known mental health problems, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained signi?cant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90 95% CI but were more likely to test positive for alcohol 1.2, 95% CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (N 9,407) and approximately 85% of those without known mental health problems (N 9,357) in 27 states (Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% CI 0.7-0.8) than were persons with known mental health problems. Whereas two thirds of decedents with known mental health problems had a history of mental health or substance use treatment just over half were in current treatment. Decedents without known mental health problems had a significantly higher likelihood of any relationship problemfloss than did those with known mental health problems speci?cally intimate partner problems (30.2% versus (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than were those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminal-legal problems (10.7% versus or evictioni?loss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobi??nancial problems were commonly experienced among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.8%, respectively). Decedents without known mental health problems had significantly lower odds of recent release from any institution 0.5, 95% CI but those who were recently released were significantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcohol/substance treatment) extra prepontion. decedents with known mental health problems who were recently released from an institution y? Level Comment Ihave already 46.7% fjwere released l'rorn corrected. 5? Decedents without known mental health problems were signi?cantly less likely to have a history of suicidal idcat'ion or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%. respectively}. Suicide intent was discloaed by 22.4% and 24.5% of persons without and with known mental health problems? respectively. Conclusions and Comments During 999?20 61 suicide rates increased signi?cantly in 44 states. and 25 states experienced increases ofmore than 30%. Rates increased signi?cantly among males in 34 states, and females. in 43 states. This ?nding is consistent with prior research showing a decreasing gender gap in male-female suicide rates during 1999-2014 Additional research into the speci?c causes of these trends is necessary. Data from the 27 states participating in provides important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identification ofsuieidal persons, treatment of mental health problems and prevention of reattempts. This study found that more than halfof suicide decedents in did not have a known mental health problems. indicating that additional focus on non-mental health factors, further upstream. is essential to a public health approach ti). This group suffered more from relationship problems and other life stressors such as criminal-legal matters, evictionfloss of home. and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as jobf?nancial andfor physical health problems. These ?ndings point to the need to both help persons manage and prevent the conditions associated with mental health problems in the first place, and to support persons with known mental health problems to decrease their risk of poor outcomes (11). Two thirds of this group had a history of any mental health andfor substance use treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. Additionally, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (12). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem- solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicab?mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the explain this strategy, eg. the meaning of {media in order to prevent suicide contagiorj Some states. such as Colorado. are planning to I, Cnmmenl ?ease elaborate to implement such a comprehensive approach to suicide prevention ?safe reporting" and ?suicide contagion. Please revise or respond. Level 2 The ?ndings in this report are subject to at least three limitations. In the state-level analysis. rankings for four states (Maryland. Massachusetts. Rhode Island. and Utah] might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward). or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second. is not yet nationally representative; the 27 states included represent 49.6% of the population Finally, abstractors data are limited to information contained in investigative reports. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies including more in?depth interviews with neat?of?kin often identi?v greater attributions to mental disorders however. many methodological variations across studies exist It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were inlknown. and hence underreported by key informants. Nonetheless. the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range oi factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: or Technical Package ofPoii'cies. Program-is. and Practices (12) to better understand their suicide problem. prioritize evidence-based comprehensive suicide prevention. and save lives. ummenl IOADSI: Some references seem incomplete. lacking hyperlinks or detailed sourcefjournal information. {Comment Deb. I have already Acknowledgments corre?edfcompleted all the references. Robert Anderson, 1y Hedegaard. and Margaret Warner. Division of Vital Statistics, National Center for Health Statistics, CDCCon?ict of Interest ii II it No eonlhets of interest were reported. ii a it lDivision of Violence Prevention. National Center for ln_iury Prevention and Control. CDC: :1 ot?Analysis. Research. and Practice integration. National lCenter for injury Prevention and Control. CDC. Corresponding author: Deborah M. Stone. dstone3@cde.gov . 770-488-3942. U.- nu-u. [leferenee CDC. Web-based Injuryr Statistics Query and Reporting System Atlanta. GA: 1. National Center for injury Prevention and Control. Retrieved March 15. 2018. In. Ivey~Stephenson AZ. Crosby AE. Jack SPD. Ha ileyesus T. Krosnow?Sedaeea Suicide 2. Trends Among and Within Urbanization Levels by Sex. Racer'Ethnicity. Age Group. and Mechanism of Death?United States. MMWR Surveill Summ 3. Curtin SC. Warner M. Hedegaaro H. increase in suicide in the United States. [999?2014. In: US Department of Health and Human Services. CDC. National Center for Health Statistics Hyattsville. MD: 2016. 4. Koehanek K. Murphy S. Xu J. Arias E. Mortality in the United States, 2'316. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 10 10. 11. 12. 13. 14. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. Zalsman G, Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10?year systematic review. Lancet Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. Crepeau?Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2010; 14:24?34. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. Caine ED, Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017:1njuryprev-2017-042366. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 11 Summary What is already known about this topic? In 2016, nearly 45,000 deaths were caused by suicide in the United States. What is added by this report? During 1999-2016, suicide rates increased in nearly every state, including 380% increases in 25 states. Mental health problems often contribute to suicide; however, 2015 data from the National Violent Death Reporting System (27 states) indicate 54% of suicide decedents were not known to have such problems. Other contributors included relationship, substance use, health, and jobx??nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in CDC ?s Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025. 12 TABLE 1. Selected demographic and descriptive characteristics of suicides aged 210 years with and without known mental health problems National Violent Death Reporting System, 21' states,* 2015 Known mental health pro ble n11 {n No known mental health problem Adjusted Characteristic Total {n 20,443} 9,407} {n 11,039} Chi?Square {95% Cl} {95% Cl} Sex Male 15,?02 3,439 {33.3} 9.233 {33.3} [340.01 2.3 Female 4,?44 {23.2} 2,933 {31.2} 1.305 {13.4} ps0.01 0.4 Age (yrsl** 10?24 2,304 1,211 {12.9} 1.593 {14.4} ps0.01 1.1 25?44 3,453 {31.3} 3.033 {32.3} 3.420 {31.0} ps0.05 0.9 45?34 3,320 {40.3} 3.393 {35.3} ps0.01 0.3 235 3,433 1,340 {14.2} 2,123 {19.3} ps0.01 1.4 Racelethnicity White, non-Hispanic 1?,102 {33.3} 3,135 {33.3} 3.93? {31.0} p<0.01 054103?01} Black. non-Hispanic 1,223 411 31? p<0.01 American lndianl?Alasi-ca Native. non- 3?3 112 233 {Jr-10.01 2.0 Hispanic Asian, non-Hispanic 5?5 235 341 p<0.05 1.2 Hispanic 1.093 43334.9} 333l5.?} 1.2 {1.0?1.3 Other 33 21 45 p<005 1.3 Extended demographics Ever served in military?rr 3,429 1,354 {15.3} 2.0?5 {20.1} ps0.01 1.4 1.1 {1.0?1 Homeless 240 104 133 1.1 1.2 Incident ije Single suicide 20.033 {93.2} 9,313 {99.1} 10,?45 0.3 0.4 Homicide followed by suicide 319 34 255 $10.01 3.5 2.9 Multiple suicides 34 25 39 1.3 1.3 Method Firearm 9.909 {43.5} 3,321 {40.3} 3.033 {55.3} p<001 1.3 1.3 5.90? {23.9} 2,940 {31.3} 2.93? {23.9} {340.01 0.3 0.3 Poisoning 3,003 1,331 {19.3} 1.142 {10.4) {340.01 0.5 0.3 Substance class causing deatl?f?? Other (9.9., over-the-counter} 1,021 {34.0} 333 {35.3} 355 {31.1} p<0.01 0.3 0.9 Opioids 944 {31.4} 303 333 {29.4} 0.9 0.9 {0.3?1 Antidepressants 300 {23.3} 344 {34.3} 153 ps0.01 0.3 0.3 Benzodiazepines 324 {20.3} 433 {25.1} 153 p<0.01 0.5 0.5 219 195 {10.5} 24 {3:001 0.2 0.2 Other 1.595 ?30 315 p<0.05 0.9 0.9 Toxicology Results Any toxicology testing 13,31? {35.1} 3,353 3.359 {30.3) p<001 0.3 Positive for 21 substance? 9,913 5,192 4.?21 ps0.01 0.3 Substance Alcohol Tested 10.950 {53.3} 5,409 5,541 {50.2} p<0.01 0.3 Positive 4,442 {40.3} 2,115 {39.1} 2.32? {42.0} {30.01 1.1 1.2 Opioids 13 Tested 5,554 (41.8) 4,258 (45.3) 4,295 (38.9) ps0.01 0.8 0.3 Positive 2,2?9 (26.5) 1,238 (29.1) 1,041 (24.2) (3:20.01 0.8 0.9 Benzodiazepines Tested 3,124 4,225 (44.9) 3.893 (35.3) (3-1001 0.7 0.7 (Off?0.3) Positive 2,454 (30.3) 1,539 (33.3) 325 (21.2) ps0.01 0.4 0.5 (0.5-0.5) Cocaine Tested (39.0) 3,866 (41.1) 4,112 p<0.01 0.9 0.9 (0.94 .0) Positive 499 (5.3) 215 (5.5) 233 (5.9) 4p<0.05 1.2 (1.0?1 .5) Amphetamines Tested N315 3,596 (39.3) 3.919 (35.5) p<0.01 0.9 0.9 Positive 235 37'5 (10.2) 350 (9.2) 0.9 1.0 (0.8?1 .1) Mariiuana Tested 5,569 (32.1) 3.12? (33.2) 3,442 (31.2) ps0.01 0.9 0.9 Positive 1,4?1 (22.4) T10 251 (22.1) 1.0 0.9 Antidepressants Tested 5,425 (25.5) 3,103 (33.0) 2,322 (21.0) ps0.01 0.5 0.5 Positive 2,214 (40.8) 1,735 (55.9) 429 (20.5) ps0.01 0.2 0.2 Abbreviation: CI con?dence interval. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronerimedical examiner or law enforcement reports. it Odds ratio re?ects the risk among those without known mental health problem relative to those with known mental health problem. i Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex, race and ethnicity. Known mental health problem was used as the reference group. Decedents were aged :10 years, as per standard in the suicide prevention literature. it Denominator is decedents aged 213 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. 14 Comment Please consider TABLE 2. Circumstances Preceding Suicide among Deoedents Aged :10 years with and without known mental health prohlems National Violent Death Reporting System. 2? revising or elaborating on what is meant states,? 201 5 Known mental in this context by, ?unknown" to health problem?. problem. no. Chi? Adjusted 0R1 1' it from, no known E.g. "Dt Characteristics Total no. {114.} Square om 15511. on @511. on} reported? or ?not documented" or other. Suicide with known circumstances 13.764 (91.8} 9.40? (100} 9.35? {34.8} r? 3 Mental Health Any Current Diagnosed Mental Health Problem" 1' IDTE Ii Anxiety disorder 1.579 (16.3} 1* Bipolar disorder 1.431 {15.2} Schizophrenia 509 PTSD 424 (4.5) - - 226 - - Unknown 150 Jr Current depressed mood 3.9621421} 3.0715 (32.9} 0.7 Substance Problems Any Current substance problem 5,319 {23.3} 2,315 {31.6} 2,343 {23.0} p<0.01 01' Aleahel problem 3.268 (11.4} 1.362 (13.31 1.4031150) p<0.01 or 0.7 Other substance problem 3,034 {16.4} 1.763 {13.8} 1,316 {14.1} Treatment Current mental healtl'll'substanee abuse 5.141 50?? {54.0} 64110.?11 0.01 {0.01-0.01} {3.01 (0.01?0.01) treatment Ever treated for mental healthisutlstance 631? (35.3] 6.323 {612} 394 {41.2} ps??i 0.02 {0.02%.02} 0.02 {0.02?0.03} problem Relationship Problemleoss Any relationship prohieml'loss 1.9401424) 3.?261396} 4.222 {45.1} p<0.01 1.3 1.3 Intimate partner problem 5,093 2.210 (24.1} 2.323 (30.2} p50% observed in 25 states (Supplementary Table; (Figure). Modeled suicide rate trends indicated significant increases in 44 states, among males (34 states} and Females (431 states). as well as for the United States overall (Supplementary Table; Alaska, Arizona, Colorado, Connecticut. Georgia, Hawaii, Kansas, Kentucky, Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New jersey. New h?lesico. New York. North Carolina, Ohio, Oklahoma. Oregon. Rhode Island, South Carolina. Utah, Vermont. Virginia. and Wisconsin. US Department of Health and Human Servicestenters for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 (a I Increase: 33%?58% I Increase: ncrease119%~30% ['22 Increase: 6%-18% Decrease: 1.0% Per 100,000 population, age?adjusted to the 2000 0.5. standard population. Nationally, the model?estimated average annual percentage change For the overall suicide rate was an increase of} .5?30. By sex, estimated national rate trends Further indicated signi?cant average annual percentage change increases For males . and Females (Supplementary Table; cdci53785). Suicide dec 3 without known mental health conditions ?1 1,039; compared with those with known mental health conditions ?9,407: 45%} #127 states. Whereas all dece- dents were predominately male (Table l) and non- Hispanic white those without known mental health conditions. relative to those with mental health conditions, were more likely to be male (83.6% versus 68.8%; odds ratio 2 2.3, 95% CI 2.2?2.5} and belong to a racial/ethnic minority (OR range Suicide decedents without known mental health conditions also had significantly higher odds oiperpetrating homicide followed by suicide 2.9, 95% Cl Among decedents aged 218 years, 20.1% of those without known mental health conditions and 15.3% of? those with mental health conditions had miserved in the US. military or were serving at the time ofdeath. Whereas firearms were the most common method of suicide overall decedenrs without known mental health conditions were more likely to die by firearm and less likely to die by or poisoning than were those with known mental health conditions 31.3%, and 19.8%, respectively). These dil?Terences remained signi?cant in the adjusted models. MMWR June 8,2018 Vol.6? No.22 5 PROOF PROOF PROOF PRUOF Morbidity and Mortality Weekly Report TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health condition1 health condition Adjusted 0R1 Characteristic 20,446) in 9,40?) (n 1 1,039] Pavalue (95% Cl] (95% Cl] Sex Male 15.702 (26.8] 6.469 (68.3] 9.233 (33.6] {0.01 2.3 NA Female 4,744 (23.2] 2,938 (31.2] 1,806 (16.4] <0.0'i 0.4 NA Age ins)? 10?24 2,304 (13.2] 1,211 (12.9] 1,593 (14.4] <00] 1.1 NA 25?44 6,456 (31.6] 3,036 (32.3] 3,420 (31.0] {0.05 0.9 (03?10] NA 45?64 7,713 (37.7] 3,320 (40.6] 3,398 (35.3] ?40.01 0.3 NA 265 3,463 {17.01 1,340 (14.2] 2.123 (19.3] (0.01 1.4 NA RaceJ'Ethnicity White. non~Hi5panic 1?,102 (83.6] 3,165 (86.8] 3.93? (31.0] c001 0.6 NA Black. non~HiSpanic 1328:1110] 411 81? {0.01 NA American Indianr'Alaska Native, 323 112 266 <0.01 2.0 {i 6?2.6] NA non?Hispanic Asian, non-Hispanic 235 341 (ti-.05 1.2 NA Hispanic 1,096 463 633 {0.05 1.2 (10-13] NA Other 66 21 45 {0.05 1.3 NA Extended demographics Ever served in militarviir 3,429 (17.8] 1.354 (15.3] 2,075 (20.1] - 1.1 (1 .0?1 Homeless 240 104 136 1.1 1.2 Incident type Single suicide 20,063 (98.2] 9,31 8 (99.1] 10, 245 (92.4] {0.01 0.3 0.4 (0.3-0.5) Homicide followed by suicide 31 9 64 (0.70 255 {0.01 3.5 2.9 Multiple suicides 64 25 39 NS 1.3 1.6 Method Firearm 9,909 (43.5] 3,321 (40.0] 6,038 (55.3] (0.01 1.3 {1.2?1.9i 1.6 5.907 (23.9] 2,940 (31.3] 2,96? (26.9] 0.8 0.8 Poisoning 3,003 (14.2] 1,861 (19.3] 1,142 (10.4] {0.01 0.5 061106?03] Substance class causing death?? Other over?the-counter] 1,021 (34.0] 666 (35.8] 355 (31.1] {0.01 0.3 0.9 Dpioids 944 (31.4] 608 (32.75] 336 (29.4] NS 0.9 (034.0] 0.9 (0.8?1 Antidepressants 800 (25.6] 644 (34.6] 156 (1 (0.01 0.3 (02?04] 0.3 Benzodiazepines 624 {20.8} 458 (25.1] 156 (13.7] (0.01 I15 0.5 (0.44.6) 219(13] 195 (10.5] 24 (0.01 0.2 (01?03] 0.2 (0.1 Other 1,595 280 815 ?5.0.05 0.9 0.9 See table footnotes on next page. Toxicolog},r testing was less likely to be performed For dece? dents without known mental health conditions. Among those with toxicology results. decedents without known mental health conditions were .less likely to test positive For any substance overall 0.8, 95% Cl including opioids 0.90. 95% Cl but were more likely to test positive for alcohol 1.2, 95%, CI information on circumstances surrounding, suicide were available for all decedents with mental health conditions (n 9,407] and approximately 35% otthose without known mental health conditions (11 9.357] in 27 states (Table Persons without known mental health conditions were less likely to have any problematic substance use 0.7, 95% Cl 0.7?0.8] than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history ot?mental health or substance use treatment just over halt were in treatment at the time of death. 4 MMWR June8.2018 r' Vol.6? 1 No.22 Decedent's without known mental health conditions had a signi?cantly higher likelihood of an)r relationship problem! loss titan did those with known mental health condir tions specifically intimate partner problems (30.2% versus argumentsiconilicts (115% versus and perpetrating interpersonal violence in the past month versus Decedents without known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life stressors (50.5% versus 412%) such as criminal legal problems (10.7% versus or eviction/loss ofhome versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively] crisis (a current or acute event thought to contribute to the suicide} (32.9% versus All ofthese differences remained signi?cant in the adjusted models. Physical health problems and job/?nan- cial problems were commonly contributing stressots among both persons without mental health conditions (23.2% and US Department of Heaith and Human Servicestenters for Diseasa Control and Prevention PROOF FWHDOF Morbidity and Mortality Weekly Report PROOF TABLE 1. {Continued} Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health conditionJr health condition 0115 Adjusted Characteristic 20,446} {n 9,402} {n 11,03 9} P?value {95% {95% Cl} Toxicology results Any toxicology testing 11312116531} 5,058 {70.8} 5,559 {60.3} (0.01 0.5 0.7 Positive for a} substance? 9,913 ire-n 5,192 was} 4.721 {70.9} (am or ins?or} as {or?0.3} Substance detected?i? Alcohol Tested 10,950 {53.6} 5,409 {52.5} 5,541 {50.2} (0.01 0.7" 0.0 Positive 4,442 {40.6} 2,115 {39.1} 2,322 {42.0} <0.01 1.1 1.2 Opioids Tested 13,554 {41.8} 4.258 {45.3} 4,296 {33.9} (0.01 0.8 0.3 Positive 2,229 {26.6} 1,233 {29.1} 1,041 {24.2} (0.01 0.3 0.9 Benzodiazepines Tested 3.124893} 4,226 {44.9} 3,393 {35.3} (0.01 0.7" 0.7 Positive 2,464 {30.3} 1,639 {33.3} 325 {21.2} <0.01 0.4 0.5 Cocaine Tested 2,923 {39.0} 3,866 {41.1} 4,1 12 {32.2} {0.01 0.9 0.9 Positive 499 216 283 {0.05 1.2 1.2 Amphetamines Tested 2,615 {32.2} 3,695 {39.3} 3,919 {35.5} {0.01 0.9 0.9 Positive 7'36 326 {10.2} 3150 NS 0.9 1.0 {0.3?1 Marijuana Tested 5,569 {32.1} 3,127 {33.2} 3,442 {31.2} 6.0.01 0.9 0.9 PositIVE 1,471 {22.4} 710 3?61 {22.1} NS 1.0 0.9 Antidepressants Tested 5,425 {26.5} 3,103 {3 3.0} 2,322 {21.0} {0.01 0.5 0.6 Positive 2,214 {40.3} 1,?35 {55.9} 1129 {20.6} s:0.01 0.2 0.2 Abbreviations: Ci con?dence interval; NA 2 not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. Decadent had been identi?ed as having a current diagnosis of mental health condition In coronerr?medical examiner or law enforcement reports. 5 DR reflects the risk among those without known mental health condition relative to those with known mental health condition. 1' Logistic regression was used to estimate adjusted OR with 95% C15 after controlling for age, sex, race, and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years. as per standard in the suicide prevention literature. Denominator is decedents aged 218 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedents with any toxicology testing. Denominatorfor each positive group is the number tested for the substance in that group. 15.6%, respectively} and those with mental health conditions {21.4% and 16.8%, respectively}. Similarly, among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. without known mental health conditions had signi?cantly lower odds of recent release from any institu- tion 0.5, 95% CI Among those recently released. decedents without known mental health conditions were significantly more likely than decedents with mental health problems to have been released from a correctional Facility {25.2% versus hospital (43.7% versus or other Facility. such as an alcoholisuhstancc use treatment Facility {24.2% versus 1 Among with known mental health conditions who were recently released from an institution, 46.7% were released From facilities. Decedents without known mental health conditions were significantly less likely to have a history of suicidal ideation US Department of Health and Human Servicesr?Centers for Disease Control and Prevention or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%, respectively}. Suicide intent was disclosed by 22.4% and 24.5% oi?pcrsons without and with known mental health conditions. respectively. Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increasics 3.30%. Rates increasod signi?cantly among males in 34 states. and Females in 43 states. Additional research into the specific causes of these trends is needed. Data From the 27 states participating in provide important insight into cir? cumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often MMWR June 8. 2018 Vol.6? r' No. 22 5 PROOF PROOF PROOF Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged 1:10 years with and without known mental health conditions National Violent Death Reporting System, 27 states,* 2015 Known No known mental health mental health condition?i condition 01315 Adjusted 01111 Characteristic Total no. no. P-value 195% 195% Suicide with known circumstances 18.764 191.81 9.407 1100} 9.357 134.3} ?10.01 no. Mental health Any current diagnosed mental health condition? ?1'1 7,076 175.21 WA MA MA MA Anxiety disorder ?t1 1.519 115.31 NM NM NM Bipolar disorder tr 1.431 115.21 WA WA 117A Schizophrenia ?11 509 15.41 MA MA MA MA PTSD ?11 424 14.5} NA 11 226 12.4] are WA MA Not specified #11 760 18.1} MA MA MA MA Current depressed mood0 7.033 1311.51 3.9521411} 3,1115 132.91 c001 0.7 toe-11.11 arias?11.11 Problematic substance use Any 5.3191203} 2,976 131.6} 2.343 125.0} (0.01 0.7 0.7 Alcohol 3.2681174} 1.8621118} 1.4061150} 4:001 0.7 0.710.7-0.Bi Other 3.0841164} 1.7681103} 1.3161141} <0.01 07103?08} Treatment Current mental healthr'suhstance use treatment 5.141 127.4} 5.077 {54.0} 64 10.7} <0.01 0.01 10.01~0.01} 0.01 10.01 ?0.01} Eyer treated for mental healt hrsuhstance disorder 6.71 7 135.8} 6.323 162.2} 394 14.2} {0.01 0.02 {0.02?0.02} 0.02 (0.02?0.03) Relationship problemsrloss Any relationship problemr'loss 73401424} 17261396} 42221431} <0.01311 2?1.3} 1.3 Intimate partner problem 5.0931212} 2.2701241] 2.3231302} {00141.1 3?1.5} 1.4 Perpetrator ofinterpersonal violence in past month 41412.2} 131 11.4} 283 13.0} (001211 B- 2 20116-24} 1riictim of interpersonal violence in past month 3410.4} 5310.6] 31 10.3} <0.05 6104?09} 11.010.54.21 Family relationship problem 1.671 13.9} 373 19.3} 295 13.1.0 {0.9?1 Other relationship problem inonintimatel 40312.1} 20212.1} 201 12.1} NS 1.0 10.8? 1 2} 1.1 Argument orcon?ict inot specified} 2.9141155} 1.278 113.61 1.6361175} <0.01 1.3 1.4 Death of a loved one {any} 1,497 13.0} 326 {13.3} 671 17.2} <0.01 0.3 10.7?11.9} 0.9 Nonsuicide death 1,181 16.3} 541' 16.9} 53415.7} c001 0.3 10.7?13.9} 0.9 Suicide of family or friend 37912.0} 217 12.3} 162 11.7} (0.01 0.7 10.64.91} 0.8 Other life stressors Any life stressor 9.171 140.9} 4.4421472} 4.7291505} (.001 1.1 11.1?12] 1.1 Recent criminal legal problem 1,588185} 58616.2} 1,0021107} (0.01 1.811.6?20} 1.7 Other legal problem 745 14.0} 373 14.0} 320 14.0} NS 1.0 108?1 1.0 Physical health problem 4.1791223} 2,012121.4} 2.1671232} 4:10.01 1 0.0?1.2} 1011.04.11 Jebrnnancial problem? 2.941 115.21 1,530 115.31 1.411 {15.51 was as {as-1.01 0.9 10.3- 1 1:11 Eviction or loss ofhome 722 13.0} 317 13.4] 405 14.3} <0.01 1.3 11. 1? 1. 5} 1.4 School problem*? 162119.91 70 {12.131 92 {21.9} NS 3102?9 1.3} 1.3 Recent release from an institutionil'i 1.412 17.6} 941 1111.21 471 15.11 ?3.111 5.11.1 4-11. 51 as lailr?PrisonfDetention facility 203 114.4} 8218.2} 121 {25.7} {0.0161234 914.5 Hospital 517136.61 311 {33.0} 206143.71} <0.01611.?3 2 0} 1.3 hospitalr'institution 469133.21 439146.71 3016.4} (0.01 0.1 10.1 1} 0.1 Other {includes alcoholrSU treatment facilities} 223115.01 109111.61 114124.21 c001 2.411.3-33} 2.511.833} See table footnotes on next page. oriented toward mental health conditions alone with regard to identi?cation of suicidal persons, treatment of mental health conditions. and prevention of teattempts. Th is study Found that approximately halfofsuicide in did not have a known mental health condition. indicating that additional focus on health Factors further upstream could provide important information for a public health approach (.161). Those without a known mental health condition suffered more From relationship problems and other 1ti such as criminalilcgal matters, cvictioniloss othomc. and recent or impending criscs. r3 MMWR June8.2018 r' I No. 22 Similarly1 persons with mental health conditions also often experienced other factors such as relationship problems and o-theelde?seresse-rs-eaeh?as johr?tinancial or physical health prob- lems that contribured to their suicide. I hese ?ndings point to the need to both prevent the circumStances associated with the onset health conditions and support persons wida known mental health conditions to decrease their risk for poor outcomes Two thirds of suicide with mental health conditions had a. history of treatment for mental health or substance use disorders, with approximately hallT in when they died. This Finding suggests the US Department of Health and Human Servicestenters for Diseaso Control and Prevention PROOF Paoor Pager Morbidity and Mortality Weekly Report PROOF TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, 27 states,? 2015 Known No known mental health mental health condition+ condition 0R5 Adjusted Characteristic Total no. no. [ii-'13} P-value (95% Cl} [95% Cl} Crisis within past or upcoming 2 5,525 {29.4} 2,444 {25.0} 3,081 {32.9} <0.01 1.4 Intimate partner problem 136835.15} 854 {34.9} 1,114 {36.2} NS 1.1 .1 Physical health problem ?#39 {13.4} 315 {12.9} 424 (13.3} NS 1 1 1-0 Criminal problem 621 [11.2) 203 418 {13.6} (0.01 1 .7 1.6 Family relationship problem 430 [18} 21203.10 218 {11} <0.05 0.8 0.9 [03?1 Job problem 354 {15.4} 191 153 0 7' 01:05?03} Suicide evantt'history Left a note 6,463 (34.5} 3,132 (33.3] 3,236 {35.1} NS 1.1 1.2 Disclosed suicide intent 4,405 123.5} 2,306 (24.5} 2,099 {22.4} com 0.9 0.9 History' of ideation 5,9901313} 3,838 (40.8} 2.152 {23.0} <0.01 0.4 0.4 History of attempts 3,732 (19.9} 2,370 {29.4} 962 {10.3} 40.01 0.3 0.3 Abbreviations: ADDEADHD attention de?cit disorderrattention deficit hyperactivity disorder; Cl 2 con?dence interval; MIA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; St] substance use. Alaska. Arizona. Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky, Maine, Maryland. Massachusetts. Michigan. Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or Iainr enforcement reports. ii DR re?ects the rislt among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for age. sex. race, and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually.r exclUsive. Therefore, sums of percentages for the dlagnosod conditions exceed 100%. Denominator includes the number of dECEdents with one or more current diagnosed mental health conditions. it Decedents with no known mental health conditions do not have mental health conditions; therefore total values are equal to the known mental health condition values. '55 Not a diagnosis. Denominator is decedents aged 2-13 years. ?it Denominator is decedents aged 10?1 El years. i? Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the pa st mo nth. Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonly occurring categories. need for additional safety supports, including broader imple- mentation of affordable and effecrivc treatment modalities, such as doctor-patient collaborative care models and proven cognitive-behavioral therapies. In addition, incrcasod access to behavioral health providers in areas is needed. as is expansion ofhcalth care that integrate physical and behavioral health, with a priority.? on suicide prevention and patient safety, especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports leg, housing stabilization pol icics, household ?nancial support); reaching coping and problem-solving skills to manage everyday stressors and prevent Future relationship problems. especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible. emotional. and social support; and identifying and better supporting persons at risk (cg, militaryr veterans. persons with physical} mental health conditions) Other strategies include creating protective environments (cg, reducing access to lethal moans among persons at risk for suicide, creating organizational and workplace policies to promote help?seeking, easing transitions US Department of Health and Human Services/Centers for Disease Control and Prevention into and out oi'wo rk for persons with mental health conditions and other life challenges), strengthening access 3nd delivery of care, supporting family and friends after a suicide, and assuring the media follow safe reporting recommendations Some states, such as Colorado. are planning to implement such a comprehensive approach to Suicide prevention The findings in this rcporrarc subject to at least three limita? tions. First, in state?level analysis, rankings for four states (Maryland, Massachusetts, Rhodc island, and Utah) might have been affected by large proportions of injuryr deaths of undetermined intent [potentially biasing reported suicide rates downward} or decreased percentages ofsuch deaths over time (potentially biasing estimated rate trends upward}. Second, is not yet rottii'lnall}r representative; the 27 states included represent 49.6% of the population Finally1 abstractors of data arc limited to information contained in investigative reports. Therefore, the extent of informant knowledge can affect data completeness and accuracy. Studies that include more interviews with next?of?kin often identify greater attributions to mental disorders however. man}r methodological variations MMWH r" June 8. 2018 Vol.6? r' No.22 Ni PROOF PROOF ROOF PROOF Morbidity and Mortality Weekly Report Summary What is already known about this tepic? In em 6. nearly 45.0iiupersons died by suicidein the United States. Mental health conditions are one ofsevereioontributurs-to. suicide. What is added by this rcport? During 1999?2915. suicide rates increased in nearly every state.- including )30% increases in25 state:-. 2015 data from 27 states indicate 54% of suicide decedent; Were not. known to have mental health conditionsGEIer-ee?eibutamded substance nee. health, and job or ?nancial Firehouse-groom; ?rt-idler;L What are the lm pllcations for public hoalth practice? A-comprehen?ve approach using prawn prevention strategies, such asthosein (IDES Technical Package for Suicide Prevention. can help ream the national goal of redudng the annual suicide Iate 20% by 2025. across studies exist It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown, and hence underrcportcd by key informants. Nonetheless, the high prevalence ofdiverse contributing circumstances among those with and without known mental health conditions suggests the importance oiaddressing the broad range offactors that contribute to suicide. Suicide is a growing public health problem. Effecrive approaches to prevent the many suicide risk factors are avail- able. States and communities can use data from RS and resources such as Preventing Suicide: ATQ ical Package ofPolicy. Programs, and Practices (12) to better under- stand suicide in their populations. prioritize evidence?based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, oily H'edegaard. Margaret Warn ct. Division of Vital Statistics. National Center For Health Statistics. CDC. Con?ict of Interest No conflicts of interest were reported. IDivision of 1lv'iolerlce Prevention. National Center for Injury Prevention and Control. CDC: 2Division of Analysis. Research. and Practice integration. National Center for injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone. 770-483-5942. MMWFI If Vol.6? I No.22 (1. 9. ill. I4. References .CDC. Web?based Iniury Statistics Query and Reporting System Atlanta, CA: National Center for Injury Prevention and Control: ZUIB. . hey?Stephenson AZ. Crosby AE, Jack SPD, T. Kresnow? Sedacca Suicide trends among and within urbanization levels by sex. raccl'ethnicity, age group, and mechanism oldesth?Unilctl States, MMWR Surveill Summ 201?;66iNo. orgfl D. 5585! Sal . Curtin SC. Warner M. Hcdegaard H. Increase in suicide in the United States. l999?2?14. NCHS data brief no 241. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. data! databricfsi db24i.pdf . Kochancit K, Murphy 5. Xu J. Arias E. Mortality in the United States. 20] NCHS data brief no 2.93. Hyattsvilie, MD: US Department of Health and Human Services. CDC. National Center For Health Statistics; 2017. . Office of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy For suicide prevention: goals and objectives for action. \Vashington. DC: US Department of Health and Human Services. Gil-lice oi: the Surgeon General: 2012. surgcungencralgovil ibraryl rionIr hill-report.de Zalsman G, Hawton K, Wasserman D, at al. Suicide prevention strategies revisited: 10-year systematic review. Lancet D. [0 1 65221 5?0366i Islsoosox .Torguson K. O?Brien A. Leading suicide prevention efforts unite to address rising national suicide rate. Washington. DC: American Foundation For Suicide Prevention: ZUIT. .Crcpeau?Hobson F. The autopsy and determination oi" child suicides: a survey of medical examiners. Arch Suicide Res. 1 American Association. Diagnostic and statisrical manual of mental disorders Washington. DC: American Association; 20H. Cainc ED. Reed J. Hindman J. Quinlan K. Comprehensive. integrated approaches to suicide prevention: pracrical guidance. Inj Prev 2017. Epub December .World Health (thaniaarit,rn. Risks to mental health: an overview of vulnerabilities and risk Factors. Geneva. Switzerland: World Health Organization: 2012. Stone DM. Holland KM. 110%; BN. Crosby AE. Davis SP. Wilkins N. Preventing suicid?uical package ot?poiicy. programs. and practice. Atlanta. GA: lepartment of Health and Human Services. CDC: 2M7. suicidetechnicalpacltagepcli' . Miincr A. Svcticic I. De Leo D. Suicide in absence ol?menral disorder? A rcvicvir of autopsy studies across countries. Int] Soc 20155954564. Pouliot De Leo D. Critical issues in autopsy studies. Suicide I. lit: Threat Behav 2006; 36: SID Wei-W US Department of Health and Human ServicesI?Centers for Disease Control and Prevention Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report Weekly/Vol. 67 i No. 22 June 8, 2018 Vital Signs: Trends in State Suicide Rates United States, 1999?201 6 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. fich; 'l?homas R. Simon l?th; Katherine A. Fowler, Pth; Scott R. Kegler, Renting Yuan, Kristin M. Holland. PliDl; Aslia Z. hey-Stephenson, I?th; Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are-just one Iactoico ntributing to suicide. Examining state?level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged :10 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data From the National Vital Statistics System For 50 states and the District of Columbia. Data From the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?2016, suicide rates increased sig C) tly in 44 states, with 25 states experiencing increases Rates increased significantly among males and females in 3 43 states, rmpectively. Fifty-four percent ofdecedents in 27 states in 2015 did not have a known mental health condition. Among decedents with available information, several circumstances were significantly more likely among those without known mental health conditions than INSIDE among those with mental health conditions, including relationship problemsi'loss (45.1% versus life stressors (SA-rage versus some), and recent! impending crises (32.9% versus but dress circumstances were common across groups. 7 QuickStats Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk1 prevent reattempts, and help friends and family members in the aftermath of a suicide. Continuing Education examination available at LLS. Department of Health and Human Services Centers for Disease Control and Prevention H.313: :1 a. or, F: a Morbidity and Mortality Weekly Report Introduction in 2016. nearly 45,000 suicides (15.6l100,000 population [age-adjusted? occurred in the United States among persons aged :10 years From 1999 to 2015, suicide rates increased among both sexes, all taciali ethnic groups, and all urbanization levels (2, Suicide rates have also increased among persons in all age groups <75 years, with eh-alasgest-pe-EeeH-t increase from 13.2 per 100, 000 persons [1999] to 19.2 per 100, 000 [2016]} and the greatest number ofsuicides {232,103 from 1999 to 2016) Suicide is the 10th leading cause of death and is one oi. just three leading causes that are increasing {Lei}. in addition. rates of emergency department visits for non?ital self-harm. a main risk Facror For suicide, increased 42% From 2001 to 2016 Together, suicides and self-harm injuries cost the nation approximately $6$Lbillion in direct medical and work loss costs The National Strategy for Suicide Prevention calls for a public health approach to suicide prevention with eil'orts spanning multiple levels (individual, Familyirclationship. community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by ansr single Factor. but rather, is determined by multiple factors. Despite NSSP guidance, suicide prevention largely Focuses on mental health conditions alone by identifying suicidal persons, provid- ing treatment for mental health conditions, and preventing teattempts In addition to mental health conditions and prior suicide attempts, other contributing circumstances include social and economic problems, access to lethal means substances, ?rearms] among persons at risk, and poor coping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach. the national goal, estab- lished by the National Action Alliance of Suicide Prevention and the American Foundation For Suicide Prevention, oFI'educ? ing the annual suicide rate 2000 by 2025 (7). To assist states in achieving this goal, CDC analyzed state?specific trends in suicide rates and assessed the multiple contributing Factors to suicide; this report presents options For m-ul-t?i?levelfomprehen? sive suicide prevention J?based on the best available evidence. Methods Suicide rates were analyzed for persons aged 210 years because determining Suicidal intent in younger children can be difficult Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (International Clari?cation afDr?rearer. Tenth Revision, underlying?cause?ofdeath codes Age? specilic population estimates were obtained From US. Census Bureaui'National Center For Health Statistics bridgedrrace population data releases. National and state-level suicide rate estimates were cal- culated For six consecutive 3-year aggregate periods span- ning 1999?2016 (1999-2001; 2002?2004: 2005-2007; The some series of publications is published by the Centre: For Surveillance. Epidemiology, and Laboratory Services. Centers For Disease Control and Prev?ition US. Department of Health and Human Services, ?tlant?a, GA 30329-4027. Suggested citation: [Amber names; ?rst three. then at 1L. if more than six} [Report title]. MMWR Morb Mortal Wkly Rep 2018;67rlinclusiva page numbers], Centers for Disease Control and Prevention Robert R. Redfield. MD, Director Anne Schuchat. MD. Foamy-1104,0149 Dirmar Les-lie Dauphin, Acting Annrim Diwrmr?r Stir-nee Jerome Conn. MD. Dim-Mr, [wire :fSa'enrr Quail? Chesley L. Richards. MD, MPH. Deputy Dimmfiir Pablo Health Eros-nae Services Michael F. ladetnaico. MD. MPH. Dirt-tron Center?r Epidemial'a? drdiehemmq Services MMWR Editorial and Produetion Staff {Weakly} Charlotte K. Kent. MPH. Acting Edirnrm Chief: Executive Editor Jacqueline Gindler, MD. Edam Mary Dott. MD, MPH, (Julius ?elder Tcresa F. Rutledge. Managing Editor Douglas W. Weather-wait. Lard Tannin! Wm?E?tsr Glenn Damon. Soumya Dunwiarrh. Teresa M. Hood. MS. mantra! W?rerJEd?irarr Martha F. Boyd, Lead Visual In?rmatian Sp?'iah?st Maureen A. Liz-ally. Julia C. Martinroe. Smphen R. Spriggs. Tong Yang. Wired {af?rmation Specialise 'Quang M. Donn. MBA. H. King. Terraye M. Starr. Mona Yang, In?rmerian T?bneiogy sort-rarer; MMWR Editoriai Beam Timothy ii Jones. MD. Chair-mm Matthew L. Boulton. MD. MPH Virginia A. Gains. MD Katherine Lyon Daniel. Jonathan E. Fielding, 1MB. MPH, NIBA David Fleming. MD hJ MMWR June-3.2018 r' Vol.6? I No.22 William E. Helperin. MD. MPH King K. Holmes. MD, Robin lkrda. MD. MPH Rims F. Khahbaa. MD Meadows. MSN. RN Jewel Mullen. MD. MPH. MFA jeEfNieclerdeppe. Patricia Quinlisk, MD. MPH Patrick L. Remington..MD. MPH Carlos Roig. MS. MA 1\Xi'illitu'n Roper. MD. MPH William Scha?iner. MD US Department of Health and Human ServicesiCenlers for Disease Control and Prevention PROOF PROOF PROOF PROOF Morbidity and Mortality Weekly Report 2000?2010; 2011?2013; and 201%2016). Rate estimates were age-adjusted to the U.S. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms oFaverage annual percentage changes. Characteristics oF persons aged 210 years who died by sui- cide, with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 27 srates* with complete data participating in CDC's National Violent Death Reporting System in 2015. de?nes mental health conditions as disorders and syn- dromes listed in the Diagnostic andSren?rritm?Momml Disorders, Edit-ion with the exception onlcohol and other substance use disordess, which are captured separately in NVD RS aggregates data From three primary data sources: death certi?cates. coroncri' medical examiner reports (including toxicology), and law enforcement reports. A range 0F circumstances (relationship problems, liFe stressors, and recent or impending crises) have been identi?ed as potential risk Factors For suicide. Circumstances captured by are those identified . . Decedents could have experienced multiple circumstances. Decedents with and without known mental health conditions were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios (aORsi with 95% con?dence intervals (Cls), controlling For age group, sex, and raiser" ethnicity. Results The most recent overall suicide rates (representing 2014? 2016) varied FourFold, From 6.9 (District oF Columbia) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table; Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the srudy period}, with absolute increases ranging From 0.8 per 100,000 {Delaware} to 8.1 (Wyoming). Percentage increases in rates ranged From 5.9% (Delaware) to 57.0% (North Dakota), with increases >30% observed in 25 States (Supplementary Table; (Figure). Modeled suicide rate trends indicated significant increases in 44 states. among males (34 states} and Females (431 states}. as well as For the United States overall (Supplementary Table; Nationally, the Alaska. Arizona, Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky, Maine. Maryland. Massachusetts. Michigan. Minlusota. New Hampshire. New jersey. New Mexico. New York, North Carolina, Ohio. Oklahoma, Oregon. llhode Island. South Carolina, Utah, Vermont, rVirginia. and \Wisconsin. US Department of Health and Human Services/Centers for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 .r W: a 2 . I Increasez3B%?SB% I ncrease119%?30% Increase: sari-18% El Decrease: 1.0% Per 100,000 population, age?adjusted to the 2000 U5. standard population. model?estimated average annual percentage change For the overall suicide rate was an increase 0F 1 By sex, estimated national rate trends Further indicated signi?cant average annual percentage change increases For males and Females (Supplementary Table; cdcf53785). Suicide decedents without known mental health conditions (J 1,03% were compared with those with known mental health conditions (?3,407) in 27 states. Whereas all decedents were predominately male (Table and non?Hispanic white those without known mental health condi? tions, relative to those with mental health conditions, were more likely to he male (83.6% versus 63.3%; odds ratio 2.3, 95% CI 2.2?2.5) and belong to a raciali'ethnic minority (OR range Suicide decedents without known mental health conditions also had significantly higher odds onerpetrating homicide Followed by suicide 2.9, 95% CI Among ad-ui-t-decedents aged 218 years, 20.1% oF those without known mental health conditions and 15.3% oF those with mental health conditions had ever served in the U.S. military or were serving at the time oF death. Whereas Firearms were the most common method oF suicide overall wilds-and without?kmental health conditions; t-ions; were more likely to die by Firearm (55 vessels-407%} and less likely to die by (26.9% or poisoning (10.4% These diFFerences remained significant in the adjusted models. MMWR June 8,2018 Vol.6? No.22 3 PROOF TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental PROOF health conditions National Violent Death Reporting System, 22 states,?IE 2015 PROOF Morbidity and Mortality Weekly Report Known mental No known mental Total health conditiorIJr health condition 10115 Adjusted Characteristic 20,446} {n 9,402} {n 11,03 9} P?value {95% {95% Cl} Sex Male 15,202 {76.8} 6,469 {68.3} 9,233 {83.6} <0.01 2.3 NA Female 4.2441232} 2.9331312} 1.8061164} (0.01 0.4 NA Age 10?24 2,804 {13.2} 1,211 {12.9} 1,593 {14.4} {0.01 1.1 NA 25?44 6,455 {31.6} 3,035 {3 2.3} 3.420 {31.0} {0.05 0.9 NA 45-64 2,213 {3 2.2} 3,820 {40.6} 3,898 {35.3} {0.01 0.8 NA 265 3,468 {12.0} 1,340 {14.2} 2,128 {19.3} (0.01 1.4 NA Fla cer' Ethnicity White, non?Hispanic 12,102 {33.6} 8,165 {86.8} 3,932 {81.0} {0.01 0.6 NA Black, non?Hispanic 1,228 411 812 <0.01 1.2 NA American Indianmlaska Native. 32311.3} 112 206 <0.01 2.0 {1 NA non-Hispanic Asian, non-Hispanic 526 235 341 ?10.05 1.2 NA Hispanic 1,096 46314.9} 533 c005 1.2 {1 .0?13} NA Other 56 21 10.2} 4510.4} {0.05 NA Extended demographics Ever served in militaryiT 3,429 {12.8} 1,354 {15.3} 2,025 {20.1} {0.01 1.4 1.1 Homeless 240 104 136 1.1 1.2 Incident type Single suicide 20,063 {93.2} 9,313 {99.1} 10,745 {92.4} {0.01 0.3 0.4 Homicide followed by suicide 319I 64 255 <0.01 3.5 2.9 Multiple suicides 64 25 39 NS 1.3 1.6 Method Firearm 3,821 {40.6} 6,088 {55.3} {0.01 1.6 5,902 {28.9} 2,940 {31.3} 2,962 {26.9} <0.01 0.8 0.8 Poisoning 3.003 {14.21 1.361 {19.3} 1.142 {10.4} <0.01 0.5 061015?021 Substance class causing death5 Other over-the?counter} 1,021 {34.0} 666 {35.8} 355 {31.1} {0.01 0.8 0.9 Opioids 944 {31.4} 608 {32.2} 336 {29.4} NS 0.9 0910.84.11: Antidepressants 800 {26.6} 644 {34.6} 156 {13.2} {0.01 0.3 0.3 Benzodiazepines 624(208} 468 {25.1} 156 {13.2} <0.01 0.5 0.5 219 195 {10.5} 24 {0.01 0.2 {0.1 0.2 Other 1,595 280 815 <0.05 0.9 0.9 See table footnotes on next page. Toxicology testing was less likely to be performed For dece? dents without known mental health conditions. Among those with toxicology results, decedents without known mental health conditions were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90, 95% CI but were more likely to test positive for alcohol 1.2, 95%, CI information on circumstances surrounding suicide were available for all decedents with mental health conditions {11 9,407} and approximately 35% of those without known mental health conditions (1?1 9,357} in 27 states (Table 2). Persons without known mental health conditions were less likely to have any-substance usedi-se-rel-ers 0.2, 95% C1 0.7?0.8) than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment iust over half Were in treatment at the time ofdeath. =1 MMWR June8,2018 r' 1 No. 22 without known mental health conditions had a signi?cantly higher likelihood of any relationship problem.i loss than did those with known mental health condi- tions specifically intimate partner problems {30.2% versus argumentsiconllicts {12.5% versus and perpetuating interpersonal violence in the past month versus Decedents withom known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life Stressors {34%1/0 versus 494E411 such as criminal legal problems {10.7% versus or eviction/loss oihomc versus and were more likely to have had a recent or impending {within the preceding or upcoming 2 weeks, respectively} crisis {a current or acute event thought to contribute to the suicide} (32.9% versus 26.01113}. All ofthese differences remained significant in the adjusted models. Physical health problems and jobi?finan- ciai problems were commonly contributing stressors among both persons without mental health conditions (23.2% and US Department of Heaith and Human Servicestenters for Diseasa Control and Prevention PROCH3 PHOCH313RUOF Morbidity and Mortality Weekly Report PROOF TABLE 1. {Continued} Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health conditionJr health condition 0R5 Adjusted Characteristic 20,446} in 9,402} {n 11,03 9} P?value {95% {95% Cl} Toxicology results Any toxicology testing 13,312 {65.1} 5,058 {70.8} 5,559 {60.3} (0.01 0.5 0.7 Positive for 21 substance? 9,913 ir4.4i 5,192 {ran} 4.721 {70.9} (not a? as {or?as) Substance detected?i? Alcohol Tested 10,950 {53.6} 5,409 {52.5} 5,541 {50.2} (0.01 0.7" 0.310.7?00} Positive 4,442 {40.6} 2,115 {39.1} 2,322 {42.0} <0.01 1.1 1.2 Opioids Tested 8,5 54 {41.8} 4.258 {45.3} 4,296 {33.9} (0.01 0.8 0.3 Positive 2,279 {20.6} 1,233 {29.1} 1,041 {24.2} (0.01 0.3 0.9 Benzodiazepines Tested 3,124 {39.2} 4,226 {44.9} 3,393 {35.3} (0.01 0.7" 0.7 Positive 2,464 {30.3} 1,639 {33.3} 325 {21.2} ?0.01 0.4 0.5 Cocaine Tested 7,928 {39.0} 3,866 {41.1} 4,1 12 {32.2} {0.01 0.9 0.9 Positive 49916.31 2165.61 283 {0.05 1.2 1.2 Amphetamines Tested 7,515 {37.2} 3,695 {39.3} 3,919 {35.5} {0.01 0.9 0.9 Positive 236 326 {10.2} 3150 NS 0.9 1.0 {0.3?1 Marijuana Tested 0,569 {32.1} 3,127 {33.2} 3,442 {31.2} {0.01 0.9 0.9 PositIVE 1,421 {22.4} 710 {22.2} 261 {22.1} NS 1.0 {0.9?1 0.9 Antidepressants Tested 5,425 {26.5} 3,103 {3 3.0} 2,322 {21.0} {0.01 0.5 0.6 Positive 221401021} 1,?35 {55.9} 1129 {20.6} s:0.01 0.2 0.2 Abbreviations: Ci con?dence interval; NA 2 not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identi?ed as having a current diagnosis of mental health condition In coronerr?medical examiner or law enforcement reports. 5 DR reflects the risk among those without known mental health condition relative to those with knoWn mental health condition. 1' Logistic regression was used to estimate adjusted OR with 95% C15 after controlling for age, sex, race, and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. Denominator is decedents aged 218 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedent's with any toxicology testing. Denominatorfor each positive group is the number tested for the substance in that group. 15.6%, respectively} and those with mental health conditions [21.4% and 16.8%, respectively). Similarly, among all persons with recent crises. intimate partner problems were the most common types and did not differ by group. without known mental health conditions had significantly lower odds of recent release from any institu- tion 0.5, 95% CI Among those recently released. without known mental health conditions were significantly more likely than with mental health problems to have been released from a correctional Facility (25.2% versus hospital (43.7% versus or other Facility. such as an alcohollsuhstancc use treatment Facility [24.2% versus 1 Among with known mental health conditions who were recently released from an institution, 46.7% were released From facilities. without known mental health conditions were significantly less likely to have a history of suicidal ideation US Department of Health and Human Servicesr?Centers for Disease Control and Prevention or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%. respectively}. Suicide intent was disclosed by 22.4% and 24.5% oi?pcrsons without and with known mental health conditions. respectively. Conclusions and Comments During 1999?2016. suicide rates increased signi?cantly in 44 states, and 25 states experienced increasics 3.30%. Rates increasod signi?cantly among males in 34 states, and Females in 43 states. Additional research into the specific causes of these trends is needed. Data From the 27 sratcs participating in provide important insight into cir? cumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often MMWR June 8, 2018 Vol.6? r' No. 22 f1 PROOF PROOF PROOF Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions National Violent Death Reporting System, states,* 2015 Known No known mental health mental health condition+ condition 005 Adjusted 00'? Characteristic Total no. no. Pdualue [95% [95% Cl] Suicide with known circumstances 13,264 [91.3] 9.40? [100] 9,3 5? [34.3] <0.01 NM MA Mental health Any current diagnosed mental health condition? 7,076 [25.2] 7,026 [25.2] NM WA NM NM Anxiety disorder [16.3] 1,579 [16.3] NM NM NM NM Bipolar disorder 1,431 [15.2] 1,431 [15.2] NM NM NM NM Schizophrenia 509 509 WA NM NM NM PTSD 424 42401.5] WA WA WA NM ADDIADHD 226 225 WA NM. WA NM Not specified - 76018.1] 1?60 NM NM NM NM Current deraresaed ir'i 3562M 3,962 [42.1] 3,0?6 [32.9] <0.01 Sesame-prayers; Any eurree?t?selsstanee-eroblem 5,319 [23.3] 237681.13] 2.343 [25.0] (0.01 0.2 Alcohol problem 3,263 [12.4] 1,362 [19.3] 1,406 [15.0] 1:11.01 0.2 Other substancepfobleen 3,034 [16.4] 1,763 [13.3] 1,316 [14.1] {0.01 Treatment Current healthr?substance abuse treatment 5,141 [22.4] 5,02? [54.0] 64 {0.01 0.01 [0.01?0.01] 0.01 [0.01?0.01] Ever treated for mental health/substanCe ?febleni 6,21? [35.3] 6,323 [62.2] 394 40.01 0.02 [0.02?0.02] 0.02 [0.02?0.03] Relationship problemsrloss Any relationship probleml?loss 2,943 [42.4] 3,226 [39.6] 4,222 [45.1] <0.01 1.5 1.3 Intimate partner problem 5,093 [22.2] 2,220 [24.1] 2,323 [30.2] {0.01 1.4 1.4 Perpetrator of interpersonal violence in past month 41412.2] 131 283 <0.01 2.2 2.0 [1 1.5-2.4] i?Fictim of interpersonalviolence in past month 84 53 31 {0.05 0.6 0.8 Family relationship problem 1.621 873 79801.5] N5 1.0 [0.9?1 Other relationship problem [nonintimate] 40312.1] 2020.1] 201 NS 1.1 Argument or conflict [not speci?ed] 2,914 [15.5] 1,2731135] 1,636 [17.5] {0.01 1.3 1.4 Death of a loved one [any] 1,492 326 0.3 9 Nonsuicide death 1,181 64? 534 (0.01 0.3 0.9 Suicide oftamily or friend 379 211' 162 <0.01 121.710.64.91 8 Other life stressors Any lite stressor . . . . {0.01 . . . Flecent criminal legal problem 1,533 586 1,002 [10.2] <00] 1.3 1.2 Other legal problem 743 3?3 3.70 NS 1.0 1.0 Physical health problem 4.1191223] 201291.41 2.1621232] (0.01 1-1 [1.04.2] 1.0 [1.04.1] JobiFinancialproblem? 1.530 Has] 1,411 [15.51 (0.05 as [as 101 as Eviction or loss of home 222 31? 405 1.4 School problem? 162 [19.9] 1'0 [12.3] 92 [21.9] NS 1.3 Recent release from an 1,412 941 [10.2] 471 {0.01 5[0. 4? 0 0.5 Jaill'PrisonfDetention facility 203 [14.4] 82 121 [25.2] <0.016[2. 2'4 9] 4.5 Hospital 517136.15] 311 [33.0] 206 [43.2] 1.3 hospitali?institution 46933.2] 439 [46.7] 30 c001 1001?01] 0.1 Other [includes alcoholi?S?treatment facilities] 223 [15.8] 109 [11.6] 114 [24.2] 3 3] 2.5 [1 .8-33] Seetable footnotes on next page. oriented toward mental health conditions alone with regard to identi?cation of suicidal persons, treatment of mental health conditions, and prevention of reattempts. This study Found that approximately lialfofsuicide dccedents in did not have a known mental health condition, indicating that additional focus on nonmental health Factors Further upstream could provide important information for a public health approach Those without a known mental health condition suffered more from relationship problems and other life stressors such as criminalflegal matters, evictionfloss thomc. and recent or impending crises. r3 MMWR June8.2018 r' Vol.6? 1 No.22 Similarly, persons with mental health conditions also n?cn experienced Factors such as relationship problems and other 1ti stressors such as job/financial or physical health problcme'lihcse findings point to the noed to both prevent the circumstances associated with the onset of mental health conditions in the first place and support persons with known mental health conditions to decrease their risk for poor outcomes Two thirds ofsuicide decedents with mental health conditions had a history for mental health or substance use disorders, with approximately haliin treatment when they died. This Finding, suggests the need for additional safety supports, including hroadcr implementation US Department of Health and Human Seryicestenters for Diseasa Control and Prevention PROOF TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions PROOF PROOF Morbidity and Mortality Weekly Report National Violent Death Reporting System. 27 states,? 2015 Known No known mental health mental health condition1 condition 0R5 Adjusted It'llFl'n Characteristic Total no. no. {ii-'13} Pavalue (95% Cl} {95% Cl} Crisis within past or upcoming 2 weeks'll'f 5.525(2931} 2.444 {25.0} 3,081 {32.9} c001 1.4 Intimate partner problem crisis 1,968 {35.6} 854 {34.9} 1.1 14 {36.2} NS 1.1 1 .1 Physical health problem crisis 139 {13.4} 315 {12.9} 424 {13.3} NS 1 1 1-0 Criminal legal problem crisis 621 {11.2} 203 418 {13.6} (0.01 1 .7 1.6 Family relationship problem crisis 430 [18} 212 218 <0.05 0.8 0.9 {03?1 Job problem crisis 354 191 {18} 153 40.01 0 1' 0.7 Suicide evantfhistory Left a note 6.465 {34.5} 3.132 {33.3} 3.236 {35.1} NS 1.1 1.2 Disclosed suicide intent 4.405 {23.5} 2.306 {24.5} 2.099 {22.4} c0111 0.9 {0.8?1 0.9 History of ideation 5.990 {31.9} 3.838 {40.8} 2.1 52 {23.0} <0.01 0.4 0.4 History of attempts 3.732 {19.9} {29.4} 962 {10.3} ??0.01 [1.3 0.3 Abbreviations: ADDEADHD attention de?cit disordenlattention deficit hyperactivity disorder; Cl con?dence interval; WA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; Ssh: substance abuse. Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma, Oregon. Rhode Island. South Carolina. Utah. Vermont. Virginia. and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. DR re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for age. sex. race. and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditionsr which are not mutually echUsive. Therefore. sums of percentages for the dlagnosod conditions exceed 100%. Denominator includes the number of dECEdents with one or more current diagnosed mental health conditions. 1* Not a diagnosis. Denominator is decedents aged 213 years. Denominator is decedents aged 10?1 8 years. Denominator ofinstitution subgroup is decedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month- 1? Denominator of crisis subgroup is decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonly occurring categories. of affordable and effective treatment modalities. such as doctor-patient collaborative cart: models and provcn cognitive? bchavioral therapies. in addition. increased access to behavioral health providers in underscored areas is needed. as is expansion of health care systems that integrate physical and behavioral health. with a priority on suicide prevention and patient safety. especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the full range offactors contributing to suicide. Prevention strategies include strengthening economic supports leg. housing stabilization policies. household ?nancial support}; reaching coping and problem-solving skills to manage everyday stressors and prevent future relationship problems, especially early in lift?. promoting social connectedness to increase a sense of belonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk (cg, military veterans. persons with physicalf mental heal d1 con di {ions} Other strategies include creating protective environments reducing access to lethal means among persons at risk for suicide. creating organirarional and workplace policies to promote help-seeking. casing transitions into and out of work for persons with mental health conditions and other life challenges}. Strengthening access and delivery of US Department of Health and Human Servicesr?Centers for Disease Control and Prevention care. supporting Family and friends alter a suicide. and assuring media follow safe reporting recommendations Some states. such as Colorado. are planning to implement such a comprehensive approach to suicide prevention U0). The ?ndings in this report are subject to at least three limita? tions. First. in the analysis. rankings for four states (Maryland. Massachusetts. Rhoda island. and Utah) might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward} or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward}. Second, is not yet nationally representative; the 27 states included represent 49.6% of the population susgovl' Finally, abstractors of data are limited to information contained in investigative reports. Therefore. the extent of informant knowledge can affect data completeness and accuracy. Studies that include more interviews with nest?of-kin often identity greater aririburions to mental disorders however. many methodological variations across studies exist (14). It is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown. and MMWH June 8. 2018 Vol.6? No.22 'Ml PROOF PROOF ROOF PROOF Morbidity and Mortality Weekly Report Summary What is already known about this tonic? in am 6, nearly 45,000 persons died by suicide in tire-timed States. Mental health conditions marinade-team . What is added by this report? During 1999?26916. suicide rates increased in nearly every state. including 30% Increases in 25 states. 2015 data from Hittite: indicate 54%,01' suidde not known to have mental health conditions. Other contributors Included Iteration- ship, substance use, health, and inner ?nancial problems, among-Others. What. are the Im pilcations for public health practice? A-comprehemive approach using preven prevention strategies, such asthoseirr Tec'hnicai Package for Suicide Prevention, can help teach the national goal of redUclng the annual suicide rate 20% by 2025. hence underreported by key informants. Nonetheless. the high prevalence ofdivcrse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk Factors are availr able. States and communities can use data from and resources such as Preventing Suicide: aiTechnical Package ofPolicy, Programs, and Practices (I2) to better under- stand suicide in their populations, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson. Holly Hedegaard, Margaret Warner, Division of Vital Statisrics, National Center For Health Statistics, CDC. Con?ict of Interest No conflicts of interest were reported. 'Division of 1ir/riolenct- Prevention. National Center for Injury Prevention and Control. 2Division of Analysis. Research. and Practice Integration. National Center For Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone, dsmne??icdcgov. ??0+488?3942. MMWR .lune8,2018 r? Vol.6? I No.22 6. 9. ll. 14. References . CDC. Web?based Injury Statistics Query and Reporting System ntlanta, CA: National Center For Injury Prevention and Control: ZUIS. . hey?Stephenson AZ. Crosby AE, Jack SPF), Haileyesus T. Kresnow? Sedacca Suicide trends among and within urbanisation levels by sex, tacelethniciry, age group, and mechanism (ll-dtatll?U?llud States, MMWR Surveill Summ 201?;66lNo, 35-18]. orgll D. 5585! Sal . Curtin SC. Warner M. Hedegaard Increase in suicide in the United States. 1999?2?14. NCHS data brief no 241. Hyarrsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. dh24l.pdf . Kochaneit K, Murphy S, Xu J, Arias E. Mortality in the United States, 20] NCHS data brief no 2.93. Hyattsville, MD: US Department of Health and Human Services. CDC. National Center For Health Statistics; 2(317. . Office of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy For suicide prevention: goals and objectives for action. \Vashington, DC: US Department of Health and Human Services. Cll?lice oi: the Surgeon General: 2012. ibraryl Elli-report.de Zalsrnan G, Hawron K, Wasserman D, er al. Suicide prevention strategies revisited: Ill-year systematic review. Lancet 6:3:646?59. D. [0 1 51"5221 5?03I3i3l .Torguson K, O?Brien A. Leading suicide prevention eFFotts unite to address rising national suicide rate. Washington. DC: Hmencan Foundation For Suicide Prevention: ZUIT. .Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 1 American Association. Diagnostic and statisrical manual of mental disorders Washington, DC: American Association; 20H. Caine ED. Reed], Hindrnan J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: pracrical guidance. Inj Prev 2017. Epub December 20, 2131?. l?olinjuryprev?l?17?0423n6 .World Health Organization. Risks to mental health: an overview of vulnerabilities and risk Factors. Geneva. Switzerland: World Health Organization: 2012. Stone DM, Holland KM. Bartholow BN. Crosby AE. Davis SP. Wilkins N. Preventing suicide: altechnical package of policy, programs. and practice. Atlanta. GA: US Department of Health and Human Services. CDC: 2M7. suicidetechnicalpacltagepdi' . Milner A, Svcticic j. Dc- Lco D. Suicide in Lhe absence disorder? A revicvi.r oi: autopsy studies across countries. Int] Soc i 0.1 lWlU020?64012444259 Pouliot L. De Leo D. Critical issues in autopsy studies. Suicide LiFe Threat Behav US Department of Health and Human Servicestenters for Disease Control and Prevention . Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report Weekly/Vol. 67 i No. 22 June 8, 2018 Vital Signs: Trends in State Suicide Rates United States, 1999?201 6 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone. fich; Thomas R. Simon l?th; Katherine A. Fowler, Pth; Scott R. Kegler, Renting Yuan, Kristin M. Holland. Pth; Aslta Z. hey-Stephenson, I?th; Alex E. Crosby, Abstract Introduction: Suicide rates in the United States have risen nearly 30% since 1999, and mental health conditions are just one factor contributing to suicide. Examining state?level trends in suicide and the multiple circumstances contributing to it can inform comprehensive state suicide prevention planning. Methods: Trends in age-adjusted suicide rates among persons aged :10 years, by state and sex, across six consecutive 3-year periods (1999?2016), were assessed using data From the National Vital Statistics System for 50 states and the District of Columbia. Data From the National Violent Death Reporting System, covering 27 states in 20] S, were used to examine contributing circumstances among decedents with and without known mental health conditions. Results: During 1999?20 16, suicide rates increased signi?cantly in 44 States, with 25 states experiencing increases Rates increased significantly among males and females in 34 and 43 states, rmpectively. Fifty-four percent ofdecedents in 27 states in 2015 did not have a known mental health condition. Among decedents with available infon?nation, several circumstances were significantly more likely among those without known mental health conditions than INSIDE among those with mental health conditions, including relationship problemsi'loss (45.1% versus life stressors (54.2% versus 493%), and recent! impending crises (32.9% versus but these circumstances were common across groups. 7 QuickStats Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health conditions. Implications for Public Health Practice: States can use a comprehensive evidence-based public health approach to prevent suicide risk before it occurs, identify and support persons at risk1 prevent teattempts, and help friends and family members in the aftermath of a suicide. Continuing Education examination available at LLS. Department of Health and Human Services Centers for Disease Control and Prevention .7. 3?10, 0-, I canPaper PROOF Morbidity and Mortality Weekly Report Introduction in 2016. nearly 45,000 suicides (15.6f100,000 population [age-adjusted? occurred in the United States among persons aged :10 years (I). From 1999 to 2013. suicide rates increased among both sexes, all taciall ethnic groups, and all urbanization levels Suicide rates have also increased among persons in all age groups <75 years, with the largest percent increase from 13.2 per 100,000 persons [1999] to 19.2 per 100,000 [2016]) and the greatest number ofsuicides {232,103 from 1999 to 2016) occurring among adults age 45?65 years (1.5). Suicide is the 10th leading cause of death and is one oi. just three leading causes that are increasing (L4). in addition. rates of emergency department visits for non?ital self-harm. a main risk {acror For suicide, increased 42% From 2001 to 2016 Together, suicides and self-harm injuries cost the nation approximately $69 billion in direct medical and work loss costs The National Strategy for Suicide Prevention calls for a public health approach to suicide prevention with efl'orts spanning multiple levels (individual, Familyirelationship. community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single Factor. but rather, is determined by multiple Factors. Despite NSSP guidance, suicide prevention largely Focuses on mental health conditions alone by identifying suicidal persons, provid- ing treatment For mental health conditions, and preventing teattempts (6). In addition to mental health conditions and prior suicide attempts, other contributing circumstances include social and economic problems, access to lethal means substances, firearms] among persons at risk. and poor coping and problem-solving skills Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help teach. the national goal, estab- lished by the National Action Alliance of Suicide Prevention and the American Foundation For Suicide Prevention, oFreduc? ing the annual suicide rate 2000 by 2025 (7). To assist states in achieving this goal, CDC analyzed state?specific trends in suicide rates and assessed the multiple contributing Factors to suicide; this report presents options For multilevel comprehen? sive suicide prevention based on the best available evidence. Methods Suicide rates were analyzed for persons aged 210 years because determining Suicidal intent in younger children can be difficult (8). Age-specific suicide counts were tabulated based on National Vital Statistics System coded death certificate records (International Classg?imrion efDr?reatet. Tenth Revision. underlying?cause?ofdeath codes Age? specilic population estimates were obtained From US. Census Bureaui'National Center For Health Statistics bridgedrrace population data releases. National and state-level suicide rate estimates were cal- culated For six consecutive 3-year aggregate periods span- ning 1999?2016 (1999?2001; 2002?2004: 2005-2007; The Mm series of publications is published by the Center For Surveillance. Epidemiologjt and Laboratory Services. Centers For Disease Control and Prevention (CDC). US. Department of Health and Human Services, ?tlant?a, GA 30329?4027. Suggested thallium [Amber names; ?rst three. then ct aL. ifmete than site] [Rt-port title]- Morb Mortal Wkly Rep page numbers], Centers for Disease Control and Prevention Robett R. Redfield. MD. Director Anne Schuebat. MD. {Jr-impel Depend Dimmi- Les-lie Dauphin, Acting Annrim Dirm?or?r Stir-nee Joanne Conn. MD. Dim-tan [wire qucimt-e Qmig'gr Charley L. Richards. MD. MPH. Deputy Diremrjirr Pu bit: Health Srioargfr?r Services Michael F. ladetnatco. MD. MPH. Director: Center ?r Epidemiology ?ndeiromtmji Smite: MMWR Editorial and Produetion Staff {Weakly} Charlotte K. Kent. MPH. Acting Edits?): Cliff: Executive Edits:- Jacqueline Gindlet. MD. Editor Mary Dott. MD, MPH, (Julius Editor F. Rutledge. Managing Editar Douglas W. Weather-wait. Lead Tacoma! Wm?Er?tar Glenn Damon. Soumya Tetma hi1. Hood. MS. moment Wn'rtrJEd'irart Martha F. Boyd, Lend Visual In?ltration? Spec-lair}: Maureen A. Lei-ally. Julia C. Martini-ore. SmphenR. Spriggs. Tong Yang. Virtual In?r'metimt Specialise 'Quang M. Dean. MBA. H. King. Terraye M. Starr. Mona Yang, In?rmary?! Titans/95y Somalia: MMWR Editorial Board Timothy F. Jones. MD. Charmer? Matthew L. Bunkers. MD. MPH Virginia a. Gains. Mb Katherine Daniel. P110 Jonathan E. Fielding, lull). MPH, NIBA David Fleming. MD hJ MMWR June-3.2018 r' Vol.6? I No.22 William E. Halpetin. MD. Donn. MPH King K. Holmes. MD, Robin Ikrda. MD. MPH Rims F. Khabhaz. MU their; Meadows. tho, man. an Jewel Mullen. MD, MPH. MFA Patricia Quinlisk. MD. MPH Patrick 1.. Remington. MD. MPH Carlos Roig. MS. MA William L. Roper. MU. son William Scha?iler. MD US Department of Health and Human ServicesrCenlers for Disease Control and Prevention PROOF PROOF PROOF PROOF Morbidity and Mortality Weekly Report 2000?2010; 2011?2013; and 2014?2016). Rate estimates were age-adjusted to the U.S. 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same 3-year data aggregates, employing weighted least-squares regression with inverse-variance weighting. Modeled rate trends are reported in terms oF average annual percentage changes. Characteristics oF persons aged 210 Years who died by sui- cide, with and without known mental health conditions, and the circumstances surrounding the suicides were compared in the 27 sit-lies" with complete data participating in CDC's National Violent Death Reporting System in 2015. de?nes mental health conditions as disorders and syn? dromes listed in the Diagnostic cordon: risritni'Memml Disorders, Fifi/J Edition (9), with the exception of?alcohol and other substance use disorders. which are captured separately in NVD RS aggregates data From three primary data sources: death certi?cates. coroner! medical examiner reports (including toxicology), and law enForcement reports. A range oF circumstances (relationship problems, liFe stressors, and recent or impending crises) have been identi?ed as potential risk Factors For suicide. Circumstances captured by are those identified by next oFkin as having actively contribr uted to a person?s suicide. Decedents could have experienced multiple circumstances. Decedents with and without known mental health conditions were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios (aORs) with 95% confidence intervals (Cls). controlling For age group, sex, and Results The most recent overall suicide rates (representing 2014? 2016) varied FourFold, From 6.9 (District oF Columbia) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table: Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the srudy period), with absolute increases ranging From 0.8 per 100,000 {Delaware} to 8.1 (Wyoming). Percentage increases in rates ranged From 5.9% (Delaware) to 57.0% (North Dakota), with increases >30% observed in 25 States (Supplementary Table; (Figure). Modeled suicide rate trends indicated significant increases in 44 states. among males (34 states} and Females {43 states}. as well as For the United States overall (Supplementary Table: Nationally, the Alaska. Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New jersey, New Mexico. New York, North Carolina, Ohio. Oklahoma, Oregon. Rhode island, .?iouth Carolina, Utah, Vermont, Virginia, and Wisconsin. US Department of Health and Human Services/Centers for Disease Control and Prevention FIGURE. Percent change in annual suicide rate,* by state United States, from 1999-2001 to 2014-2016 4w I I ncrease119%?30% Increase: ass-13% l] Decrease: 1.0% Per 100,000 population, age?adjusted to the 2000 U.S. standard population. model?estimated average annual percentage change For the overall suicide rate was an increase 0F 1 By sex, estimated national rate trends Further indicated significant average annual percentage change increases For males and Females (Supplementary Table; cdci?53785). Suicide decedents without known mental health conditions (1 1,059) were compared with those with known mental health conditions (9,407) in 27 states. \Whereas all decedents were predominately male (Table l) and non?Hispanic white those widlout known mental health condi- tions, relative to most: with mental health conditions, were more likely to be male (83.6?3?0 versus 63.3%; odds ratio 2.3, 95% CI 2.2?2.5} and belong to a raciah?ethnic minority (OR range Suicide decedents without known mental health conditions also had significantly higher odds onerpetrating homicide Followed by suicide 2.9, 95% C1 Among adult decedents aged 2:18 years, 20.1% oF those without known mental health conditions and 15.3% oF those with mental health conditions had ever served in the U.S. military or were serving at the time oFdeath. Whereas Firearms were the mosr common method oF suicide overall and among decedents with and without mental health conditions, decedents without known mental health conditions, relative to those with known mental health condi? tions, were more likely to die by Firearm (55 versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These diFFetences remained significant in the adjusted models. MMWR June 8,2018 Vol.6? No.22 3 PROOF TABLE 1. Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental PROOF health conditions National Violent Death Reporting System, 2? states,?IE 2015 PROOF Morbidity and Mortality Weekly Report 31:11.01: Known mental No known mental Total health conditionir health condition 10115 Adjusted Characteristic 20,446} {n {n 11,03 9} P?value {95% {95% Cl} Sex Male 15,702 {76.8} 6,469 {63.3} 9,233 {33.6} (0.01 2.3 NA Female 4.144 {23.2} 2.938 {31.21 1.806 {16.4} (0.01 0.4 {0200.5} NA Age 10?24 2.3041133) 1,211 {12.9} 1.5931144} com 1.1 NA 25?44 6,455 {31.6} 3.035 {3 2.3} 3.420 {31.0} <0.05 0.9 NA 45-64 2,218 {32.2} 3,820 {40.6} 3,893 {35.3} {0.01 0.8 NA 1265 3,468i110} 1,340 {14.2} 2.123 {19.3} <0.01 1.4 NA Racef Ethnicity White. non?Hispanic 12,102 {33.6} 8,165 {36.8} 3,932 {81.0} {0.01 0.6 NA BIack,non?Hispanic 1,228 411 3179.4} <0.01 NA American Indian/Alaska Native, 310311.51} 112 11.2} 26612.4} {0.01 20116?215} NA non-Hispanic Asian, non?Hispanic S26 235 341 ?20.05 1.2 NA Hispanic 1,096154} 463 633 (0.05 1.2 NA Other 5610.3} 21 4510.4} {0.05 1.311.1-31} NA Extended demographics Ever served in military11 3.429 1.3541153} 2.0151001} {0.01 1.4 1.1 Homeless 24011.2} 104 (1.11 135 (1.31 1.1 1.2 (0.9-1 incident type Single suicide 20.063 {93.2} 9,313 {99.1} 10.745 {914} {0.01 0.3 0.4 Homicide followed by suicide 319 64 255 <0.01 3.5 2.9 Multiple suicides 64 25 39 NS 1.3 1.6 Method Firearm 9,909 {48.5} 3.321 {40.6} 6,083 {55.3} ?5.0.01 1.8 1.6 {1.54.7} 5,902 {23.9} 2,940 {31.3} 2,9671r {26.9} <0.01 0.8 0.8 Poisoning 3.003 (14.?1 1.061 {19.01 1.142 (10.41 (0.01 0.5 05106-01} Substance class causing death?? Other over-the-counteri 1.021 {34.0} 666615.81 3551:3111 (0.01 0.0 0.9 Opioids 94431.4} 608 336 {29.4} NS 0.9 0.9 Antidepressants 300 {26.6} 644 {34.6} 156 {13.2} {0.01 0.3 0.3 Benzodiazepines 624 {20.0} 4613 {25.1} 156 {13.7} <0.01 0.5 0.5 21911.3} 195 {10.5} 24 {0.01 0.210.1?03} 0.210.1?03} Other 1,595 2130 815 <0.05 0.9 0.9 See table footnotes on next page. Toxicology testing was less likely to he performed For dece? dents without known mental health conditions. Among those with toxicology results. decedents without known mental health conditions were less likely to test positive for any substance overall 0.8. 95% CI including opioids 0.90. 95% CI but were more likely to test positive for alcohol 1.2. 95%, CI information on circumstances surrounding suicide were available for all decedents with mental health conditions in 9.407} and approximately 35% of those without known mental health conditions {r1 9.357} in 27 states (Table 2). Persons without known mental health conditions were less likely to have any substance use disorders 0.37. 95% C1 0.7?0.8} than were persons with known mental health conditions. Whereas two thirds of decedents with known mental health conditions had a history of mental health or substance use treatment iust over hall Were in treatment at the time ofdeath. i MMWR June8.2018 r' v'olJEni' I No. 22 Decedents without known mental health conditions had a signi?cantly higher likelihood of any relationship problem.l loss than did those with known mental health condi- tions specifically intimate partner problems (30.2% versus {17.5% versus and perpetrating interpersonal violence in the past month versus Decedents withom known mental health conditions were also more likely than were those with known mental health conditions to have experienced any life Stressors {54.2% versus 49.7%} such as criminal legal problems {107% versus or cvictionlloss ofhomc {4.310111 versus and were more likely to have had a recent or impending {within the preceding or upcoming 2 weeks. respectively} crisis (a current or acute event thought to contribute to the suicide} (32.9% versus All olthese differences remained significant in the adjusred models. Physical health problems and job/finan- cial problems were commonly contributing stressots among both persons without mental health conditions {23.2% and US Department of Health and Human Servicestenters for Diseasa Control and Prevention PROCH3 PHOCH313RUOF Morbidity and Mortality Weekly Report PROOF TABLE 1. {Continued} Selected demographic and descriptive characteristics of suicides among persons aged 210 years with and without known mental health conditions National Violent Death Reporting System, 2? states,* 2015 Known mental No known mental Total health conditionJr health condition 0R5 Adjusted Characteristic 20,446} in 9,402} {n 11,03 9} P?value {95% {95% Cl} Toxicology results Any toxicology testing 13,312 {65.1} 5,058 {70.8} 5,559 {60.3} (0.01 0.5 0.7 Positive for 21 substance? 9,913 ir4.4i 5,192 {ran} 4.721 {70.9} (not a? as {or?as) Substance detected?i? Alcohol Tested 10,950 {53.6} 5,409 {52.5} 5,541 {50.2} (0.01 0.7" 0.310.7?00} Positive 4,442 {40.6} 2,115 {39.1} 2,322 {42.0} <0.01 1.1 1.2 Opioids Tested 8,5 54 {41.8} 4.258 {45.3} 4,296 {33.9} (0.01 0.8 0.3 Positive 2,279 {20.6} 1,233 {29.1} 1,041 {24.2} (0.01 0.3 0.9 Benzodiazepines Tested 3,124 {39.2} 4,226 {44.9} 3,393 {35.3} (0.01 0.7" 0.7 Positive 2,464 {30.3} 1,639 {33.3} 325 {21.2} ?0.01 0.4 0.5 Cocaine Tested 7,928 {39.0} 3,866 {41.1} 4,1 12 {32.2} {0.01 0.9 0.9 Positive 49916.31 2165.61 283 {0.05 1.2 1.2 Amphetamines Tested 7,515 {37.2} 3,695 {39.3} 3,919 {35.5} {0.01 0.9 0.9 Positive 236 326 {10.2} 3150 NS 0.9 1.0 {0.3?1 Marijuana Tested 0,569 {32.1} 3,127 {33.2} 3,442 {31.2} {0.01 0.9 0.9 PositIVE 1,421 {22.4} 710 {22.2} 261 {22.1} NS 1.0 {0.9?1 0.9 Antidepressants Tested 5,425 {26.5} 3,103 {3 3.0} 2,322 {21.0} {0.01 0.5 0.6 Positive 221401021} 1,?35 {55.9} 1129 {20.6} s:0.01 0.2 0.2 Abbreviations: Ci con?dence interval; NA 2 not adjusted; NS not signi?cant; OR odds ratio. Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identi?ed as having a current diagnosis of mental health condition In coronerr?medical examiner or law enforcement reports. 5 DR reflects the risk among those without known mental health condition relative to those with knoWn mental health condition. 1' Logistic regression was used to estimate adjusted OR with 95% C15 after controlling for age, sex, race, and ethnicity. Known mental health condition was used as the reference group. Decedents were aged 210 years, as per standard in the suicide prevention literature. Denominator is decedents aged 218 years with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedent's with any toxicology testing. Denominatorfor each positive group is the number tested for the substance in that group. 15.6%, respectively} and those with mental health conditions [21.4% and 16.8%, respectively). Similarly, among all persons with recent crises. intimate partner problems were the most common types and did not differ by group. without known mental health conditions had significantly lower odds of recent release from any institu- tion 0.5, 95% CI Among those recently released. without known mental health conditions were significantly more likely than with mental health problems to have been released from a correctional Facility (25.2% versus hospital (43.7% versus or other Facility. such as an alcohollsuhstancc use treatment Facility [24.2% versus 1 Among with known mental health conditions who were recently released from an institution, 46.7% were released From facilities. without known mental health conditions were significantly less likely to have a history of suicidal ideation US Department of Health and Human Servicesr?Centers for Disease Control and Prevention or prior suicide attempts compared with those with known mental health conditions (40.8% and 29.4%. respectively}. Suicide intent was disclosed by 22.4% and 24.5% oi?pcrsons without and with known mental health conditions. respectively. Conclusions and Comments During 1999?2016. suicide rates increased signi?cantly in 44 states, and 25 states experienced increasics 3.30%. Rates increasod signi?cantly among males in 34 states, and Females in 43 states. Additional research into the specific causes of these trends is needed. Data From the 27 sratcs participating in provide important insight into cir? cumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often MMWR June 8, 2018 Vol.6? r' No. 22 f1 PROOF PROOF PROOF rumor Morbidity and Mortality Weekly Report TABLE 2. Circumstances preceding suicide among decedents aged :10 years with and without known mental health conditions National Violent Death Reporting System, 27 states,* 2015 Known No known mental health mental health condition1 condition 0115 Adjusted Characteristic Total no. no. [96] Pdualue [95% [95% Cl] Suicide with known circumstances 13,264 [91.8] 9.402 [100] 9,3 51' [34.3] <00] WA WA Mental health Any current diagnosed mental health condition?H 2,076 [25.2] 2,026 [25.2] NIA WA WA Anxiety disorder 1,579 [16.8] 1,579 [16.8] NIA NJA MIA Bipolar disorder 1,431 [15.2] 1,431 [15.2] NM WA NM WA Schizophrenia 509 509 NM NM WA NM PTSD 424 42414.Not specified 260 1'60 MA MA WA NM Current depressed moodil' 3,962 [42.1] 3,962 [42.1] 3,026 [32.9] <0.01 0.2 0.2 Substance problems Any current substance problem 5,319 [28.3] 2,926 [31.6] 2.3 43 [25.0] <00] 02' Alcohol problem 3,268 [12.4] 1,862 [19.8] 1,406 [15.0] 40.01 0.7 0.2 [0343.8] Other substance problem 3,084 [16.4] 1,763 [l 8.8] 1,316 [14.1] (0.01 0.1" Treatment Current mental abuse treatment 5,141 [22.4] 5,02? [54.0] 64 {0.01 0.01 [0.01?0.01] 0.01 [0.01?0.01] Ever treated for mental problem 6,211 [35.0] 6,323 [62.2] 394 0.02 [0.02?0.02] 0.02 [0.02?0.03] Relationship problemsiloss Any relationship problemi?loss 2,943 [42.4] 122659.61] 4,222 [45.1] Intimate partner problem 5,098 [22.2] 2,220 [24.1] 2,828 [30.2] 1.5] 1.4 Perpetrator of interpersonal violence in Past month 414 131 4] 283 <00] 2[1 .13- 2. 2] 2.0 [1.02.4] of interpersonal violence in past month 34 3.[0 6] 31 ??0.05 6014?0. 9] Family relationship problem 1.671 1373 3] 19803.5] NS 1.0] 1.0 [0.9?1 Other relationship problem [nonintimate] 403 20212.1] 201 [15001?0 1.2] 1.1 Argument or conflict [not Specified] 2,914 [15.5] 1,278 [13.6] 1,636 [17.5] {0.013il 2- 1 5] 1.4[1.341.5] Death ofa loved one [any] 1,492 826 621 0 9] 0.9 Nonsuicide death 1,181 642' 534 <00] 8.[0 2? 0 9] 0.9 Suicide of family or friend 379 211' 162 [1 <00] 13-0. 9] 0.3 Other life stressors Any life stressor 9,243 [51.9] 40750193] 5.068 [54.2] 1.2 1.1 Recent criminal legal problem 1,588 586 1,002 [10.2] <00] 1.8 1.2 Other legal problem 748 37814.0] 320 NS 1.0 1.1] 1.0 Physical health problem 2,115? [23.2] 1 .12 lobEFInancial problem? 2.941 [10.2) 1.530 [10.131 1,411 [15.01 c005 [.9011 1.0] 0910.34.01 Eviction or loss ofhome 222 31? 405 <0.013[1. 1? 1.5] 1.4 School problem? 162 [19.9] 1'0 [12.3] 92121.9] N5 1.3 Recent release from an 1,412 [2.15] 941 [10.2] sTil-[31510 4?0 5] 0.5 [13-4-05] laill'PrisonfDetention facility 203 [14.4] 82 121 [25.2] 9] 4.5 Hospital 517916.15] 311 [33.0] 2060133] <00] 6E1. 3? 2. 0] 1.3 hospitali?institution 469 [33.2] 439 [46.7] 3016.4] <00] 0.1 Other [includes alcohDIISA treatment facilities] 223 [15.8] 109 [11.6] 114 [24.2] 3. 3] 2.5 [1 .8-33] Seetahle footnotes on next page. oriented toward mental health conditions alone with regard to identification of suicidal persons. treatment of mental health conditions, and prevention of reatrempts. This study Found that approximately ltalfofsuicide decedents in did not have a known mental health condition, indicating that additional focus on nonmental health Factors Further upstream could provide important information for a public health approach Those without a known mental health condition suffered more from relationship problems and other life sttessors such as matters, cvictioniloss oihomc. and recent or impending crises. r3 MMWR :1 June 8,2018 1 Vol.6? 1 No.22 Similarly, persons with mental health conditions also often experienced nonmental health factors such. as relationship problems and ether lite stressors such as iohi?l?inancial or physical health problems. These findings point to the need to both prevent the circumstances associated with the onset of mental health conditions in the first place and support persons with known mental health conditions to decrease their rislt for poor outcomes Two thirds ofsuicidc decedents with mental health conditions had a history oftrcatrnent for mental health or substance use disorders, with approximately half in treatment when they died. This Finding, suggests the need for additional safety supports, including hroadcr implementation US Department of Health and Human SewicesiCenters for DiseaSE Control and Prevention PROOF TABLE 2. {Continued} Circumstances preceding suicide among decedents aged 210 years with and without known mental health conditions PROOF PROOF Morbidity and Mortality Weekly Report National Violent Death Reporting System, 27 states,? 2015 l3 Known No known mental health mental health condition1 condition 0R5 Adjusted Characteristic Total no. no. [as] Pavalue (95% Cl} {95% Cl} Crisis within past or upcoming 2 vveeks?ll'f 55259931} 2,444 {25.0} 3,081 {32.9} 40.01 1.4 1.4 {1 .3?1 Intimate partner problem crisis 1,968 {35.6} 854 {34.9} 1.1 14 {36.2} NS 1.1 1 .1 Physical health problem crisis 139 {13.4} 315 {12.9} 424 {13.3} NS 1 1 1-0 Criminal legal problem crisis 621 {11.2} 203 418 {13.6} (0.01 1 .7 1.6 Family relationship problem crisis 430 [18} 212 218 <0.05 0.8 0.9 {03?1 Job problem crisis 354 {15.4} 191 {18} 153 40.01 0 1' 0.7 Suicide evantrhistory Left a note 6,465 {34.5} 3,132 {33.3} 3.236 {35.1} NS 1.1 1.2 Disclosed suicide intent 4.41:15 {23.5} 2,306 {24.5} 2.099 {22.4} com 0.9 {0.8?1 0.9 History of ideation 5.990 {31.9} 3,838 {40.8} 2.1 52 {23.0} <0.01 0.4 0.4 History of attempts 3,732 {19.9} {29.4} 962 {10.3} ??0.01 [1.3 0.3 Abbreviations: ADDEADHD attention de?cit disorden'attention deficit hyperactivity disorder; Cl con?dence interval; WA not applicable; NS not signi?cant; OR odds ratio; PTSD posttraumatic stress disorder; SA substance abuse. Alaska. Arizona. Colorado, Connecticut, Georgia, Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey, New Mexico, New York. North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, Utah, Vermont. Virginia, and Wisconsin. 1 Decedent had been identified as having a current diagnosis of mental health condition in coronerfmedical examiner or law enforcement reports. DR re?ects the risk among those without known mental health condition relative to those with known mental health condition. Logistic regression was used to estimate adjusted OH with 95% after controlling for age. sex. race, and ethnicity. Known mental health condition was the reference group. Includes decedents with one or more diagnosed current mental health conditions, which are not mutually exclUslve. Therefore, sums of percentages for the dlagnosod conditions exceed 100%. Denominator includes the number of dECEdents with one or more current diagnosed mental health conditions. Not a diagnosis. Denominator is decedents aged 213 years. Denominator is decedents aged 10?1 8 years. Denominator ofinstitution subgroup is decedents with recent release from an institution. Recent release from an institution is defined as having occurred within the past month- Denominator of crisis subgroup ls decedents with any crisis within past or upcoming 2 weeks. Crises depicted here represent the most commonly occurring categories. of affordable and effective treatment modalities, such as doctor-patient collaborative carc models and provcn cognitive? bchavioral therapies. in addition, increased access to behavioral health providers in areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety especially through care transitions Comprehensive statewide suicide prevention activities are needed to address the ?dl range olifactors contributing to suicide. Prevention strategies include strengthening economic supports (cg, housing stabilization policies. household ?nancial support}; reaching coping and problem-solving skills to manage everyday and prevent Future relationship problems, especially early in life; promoting social connectedness to increase a sense if belonging and access to informational. tangible. emotional. and social support; and identifying and better supporting persons at risk (cg, military veterans. persons with physicalr? mental health con di {ions} Other strategies include creating protective environments reducing access to lethal means among persons at risk For suicide. creating organirarional and workplace policies to promote help-seeking. easing transitions into and out of work For persons with mental health conditions and other 1ti challenges}. strengthening access and delivery of US Department of Health and Human Servicesr?Centers for Disease Control and Prevention care, supporting Family and friends after a suicide, and assuring media follow safe reporting Some states, such 35 Colorado, are planning to implement such a comprehensive approach to suicide prevention U0). The ?ndings in this report are subject to at least three limita? tions. First, in the analysis. rankings for Four states (Maryland, Massachusetts, Rhodc island, and Utah) might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward} or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward}. Second, is not nationally representative; the. 27 states included 49.6% of the population Sus.govr' Faccsr?rablesc Finally, abstracrors of data are limited to information contained in investigative reports. Therefore. the extent ot? informant knowledge can attract data completeness and accuracy. Studies that include more interviews with next?oF-kin often identi?r greater artriburions to mental disorders however. many methodological variations across studies exist (14). [r is likely that some persons without known mental health conditions in the current study were experiencing mental health challenges that were unknown. and MMWH June El. 2018 Vol.6? No.22 'Ml PROOF PROOF ROOF PROOF Morbidity and Mortality Weekly Report Summary What is already known about this tragic? in 1016, nearly 45,000 persons died by Suicide in the-Uruted States. Mental health conditions can contribute to outside. What is added by this report? During 1999?2916.5uicide rates increased in nearly-every state. including 330% Increases in 25 states. 2615 data from Hittite: indicate 54% of not known to have mental health conditions. Other contributors Included relation~ ship, substance use, health, and inner ?nancial problems, among others. What. are the lm pllcations for public health practice? A-comprehemive approach using proven prevention strategies, such asthoseiri (206?s Tec'hnicai Package for Suicide Prevention, can help reach the national goal of redurlng the annual sul?de rate 20% by 2025. hence underreporred by key informants. Nonetheless. the high prevalence ofdivcrse contributing circumstances among those with and without known mental health conditions suggests the importance of addressing the broad range of Factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk Factors are availr able. States and communities can use data from and resources such as Preventing Suicide: a Technical Package ofPolicy, Programs. and Practices (I2) to better under- srand suicide in their populations, prioritize evidence-based comprehensive suicide prevention. and save lives. Acknowledgments Robert Anderson. Holly Hedegaard. Margaret Warner. Division of Vital Statisrics, National Center For Health Statisrics, CDC. Con?ict of Interest No con?icts of interest were reported. 'Division of 1ir?iolencc Prevention. National Center for Injury Prevention and Control. 2Division of Analysis. Research. and Practice Integration. National Center For injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone, ??0+483?3942. El MMWR i .ltine8.2018 Vol.6? No.22 IQ 6. 9. ll. References . CDC. Web?based Injury Statistics Query and Reporting System MSQARS). ntlanta, CA: National Center For Injury Prevention and Control: 2018. . hey?Stephenson AZ. Crosby AE, Jack SPF), Haileyesus T. Kresnow? Sedacca Suicide trends among and within urhaniaatinn levels hy sex, taceiethnicity, age group, and mechanism States, 2001?2Ul5. MMWR Surveill Summ . Curtin SC. Warner M. Hedegaard H. Increase in suicide in the United States. 1999?3114. NCHS data brief no 241. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics; 2016. db24l.pdf . Kochanelt K, Murphy S. Xu J. Arias E. Mortality in the United States, 2016. NCHS data brief no 2.93. Hyattsville, MD: US Department of Health and Human Services. CDC. National Center For Health Statistics; Zilil . Office of the Surgeon General; National Action Alliance for Suicide Prevention. 2012 National strategy For suicide prevention: goals and objectives For action. \Vashington, DC: US Department of Health and Human Services. Cll?lice oi: the Surgeon General: 2012. surgeongcneralgovll ibraryl Elli-report.de Zalsrnan G, Hawton K, Watserman D, or :11. Suicide prevention strategies revisited: Ill-year systematic review. Lancet 6:3:646?59. D. [0 1 51"5221 5?03I3i3i .Torguson K, O?Brien A. Leading suicide prevention eFForts unite to address rising national suicide rate. Washington. DC: Aime-ricer! Foundation For Suicide Prevention: 2017. .Crcpeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res American Association. Diagnostic and statisrical manual of mental disorders Washington. DC: American Association; 2013. Caine ED. Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: pracrical guidance. Inj Prev 2017. Epub December 21}. 2131?. .World Health Organization. Risks to mental health: an overview of vulnerabilities and risk Factors. Geneva. Switzerland: World Health Organization: 2012. Stone DM. Holland KM. Bartholow BN. Crosby AE. Davis SP. Wilkins N. Preventing suicide: 3 technical package of policy, programs. and practice. Atlanta. GA: US Department of Health and Human Services. CDC: 2017. suicidetechnicalpackagepdf . Milner A. Sveticic 1. De Leo D. Suicide in Lht- absence ofmental disorder? A revievi.r of autopsy studies across countries. int] Soc 0.1 124442551 . Pouliot L. De Leo D. Critical issues in autopsy studies. Suicide LiFe Threat Behav 0. 521 i suli.2006.36.5.49l US Department of Health and Human Servicestenters for Disease Control and Prevention 5.4.18 Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Pth; Keming Yuan, Kristin M. Holland, Asha Z. Ivey?Stephenson, Alex E. Crosby, Background: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple circumstances contributing to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive three-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known mental health problem. Among persons with circumstance information, several circumstances were signi?cantly more likely among those without a known mental health problems than among decedents with mental health problems, including relationship problemsfloss [45.1% vs life stressors {54.2% vs and recentfimpending crises (32.9% vs but these circumstances were common across groups. 5.4.18 Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friendsifamily after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides (15.6! 100,000 population [age?adjusted]) occurred in the United States, among persons aged 310 years (1). Between 1999 and 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide is the 10th leading cause of death and is one of just three leading causes that are increasing Additionally, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased 42% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (I). The Nations! iraiegjv for Suicide Preveniion (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyirelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factOr, but rather, is determined by multiple factors. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems and preventing reattempts (6). In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic 5.4.18 problems, access to lethal means substances, ?rearms) among persons at risk, and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the nation?s goal of reducing suicide rates 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates, assessed the multiple contributing factors to suicide, and presents options for multi-level comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for persons aged 3:10 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death ceni?cate records {International Classi?cation of Diseases 10? Revision, underlying-cause-of death codes Y87.0, U03). Age-specific population estimates were obtained from US. Census Bureaui?National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999?2016 (1999?2001, 2002?2004; 2005?2002; 2008?2018; 2011? 2013; and 2014?2016). Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics and circumstances of persons aged 3:10 years who died by suicide, with and without known mental health problems, were compared in the 27 states with complete data 3 5.4.18 participating in CDC ?5 National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certi?cates, coroner/medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known mental health problems were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% con?dence intervals (CI), controlling for age group, sex, and raceiethnicity. RESULTS The most recent overall suicide rates (representing 2014?2016) varied four-fold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (available online). Across the study period, rates increased in all states except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 per 100,000 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (available online; Figure I). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females (43 states), as well as for the United States overall (available online). Nationally, the model?estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated signi?cant increases for males (AAPC and females (AAPC (available online). Suicide decedents without known mental health problems (N 11,039) were compared with those with known mental health problems (N 9,407) in 27 states. Whereas all decedents were predominately male (Table 1) and non-Hispanic white those without known 4 5.4.18 mental health problems, relative to those with mental health problems, were more likely male (83.6% versus 68.8%; odds ratio (OR) 2.3, 95% CI 2.2-2.5) and racialfethnic minorities (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% C1 3.8). Among adult decedents 1318 years, 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever served, or were currently serving, in the U.S. military. Whereas ?rearms were the most common method of suicide overall and for decedents with and without mental health problems, decedents without known mental health problems were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% ver3us 31.3%) or poisoning (10.4% verSus 19.8%) than were those with known mental health problems. These differences remained signi?cant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall 0.8, 95% C1 including opioids 0.90 95% CI but were more likely to test positive for alcohol 1.2, 95% CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (N 9,407) and approximately 85% of those without known mental health problems (N 9,357) (Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% Cl Whereas two thirds of 5.4.18 decedents with known mental health problems had a history of mental health or substance use treatment just over half were in current treatment. Decedents without known mental health problems had a signi?cantly higher likelihood of any relationship problemfloss than did those with known mental health problems specifically intimate partner problems (30.2% versus (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminal-legal problems (10.7% versus or evictionx?loss of home versus and were more likely to have had a crisis a current or acute event thought to contribute to the suicide, within the preceding or impending, two weeks (32.9% versus All of these differences remained significant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobx??nancial problems were commonly experienced among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.8%, respectively). Decedents without known mental health problems had significantly lower odds of recent release from any institution 0.5, 95% CI but those who were recently released were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcoholisubstance treatment) 2.5 95% CI than those with a known mental health problems. Among decedents with known mental health problems who were recently released from an institution 46.7% of were released from facilities. 5.4.18 Decedents without known mental health problems were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health problems, respectively. Conclusions and Comments During 1999-2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increases of more than 30%. Rates increased signi?cantly among males in 34 states, and females, in 43 states. This ?nding is consistent with prior research showing a decreasing gender gap in male?female suicide rates during 1999?2014 Additional research into the speci?c causes of these trends is necessary. Fortunately, data from the 27 states participating in provides important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor (5), however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health problems and prevention of reattempts. This study found that more than half of suicide decedents in did not have a known mental health problems, indicating that additional focus on non-mental health factors, further upstream, is essential to a public health approach (10). This group suffered more from relationship problems and other life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as jobr??nancial andfor physical health problems. These ?ndings point to the need to both help persons manage the conditions associated with mental health problems in the ?rst place, and to support persons with known mental health problems to decrease their risk 7 5.4.18 of poor outcomes Two thirds of this group had a history of any mental health andfor substance use treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive- behavioral therapies. Additionally, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (I2). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household ?nancial support); teaching coping and problem- solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicalx?mental health problems) (I2). Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (i 2). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention (10). The ?ndings in this report are subject to at least three limitations. In the state-level analysis, rankings for four states (Maryland, Massachusetts, Rhode Island, and Utah) might have been 5.4.13 affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Finally. abstractors of data are limited to information contained in investigative reports. Therefore, the extent of infonnant knowledge can affect data completeness and accuracy. Studies including more in?depth interviews with next?of?kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package Programs, and Practices {12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard, and Margaret Wamer, Division of Vital Statistics, National Center for Health Statistics, CDC. Conflict of Interest No con?icts of interest were reported. 9 5.4.18 'Division of Violence Prevention, National Center for .Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488?3942 References 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow?Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death?United States, 2001?2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 10 5.4.18 10. 11. 12. 13. 14. Crepeau?Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2010; 14:24n34. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. Caine ED, Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017:injuryprev-2017-042366. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int .1 Soc Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 10. Summary Box (word count 1001100) What is already known on the topic? In 2016, nearly 45,000 lives were lost to suicide in the 11.3. What is added by this report? Between 1999-2016, suicide rates increased in nearly every state. Twenty?five states saw rate increases Mental health problems often contribute to suicide, however, 2015 data from the National Violent Death Reporting System (27' states) indicate that 54% of suicide decedents 11 5.4.18 were not known to have such problems. Other contributors included relationship, substance use, health, and jobf?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025. 12 Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Pth; Keming Yuan, Kristin M. Holland, Asha Z. Ivey?Stephenson, Alex E. Crosby, Background: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple circumstances contributing to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive three-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known mental health problem. Among decedents with circumstance information available, several circumstances were signi?cantly more likely among those without a known mental health problems than among those with mental health problems, including relationship problemsfloss [45.1% versus life stressors (54.2% versus and recentfimpending crises (32.9% versus but these circumstances were common across groups. Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friends and family in the aftermath of a suicide. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides (15.6! 100,000 population [age?adjusted]) occurred in the United States among persons aged :10 years (1). Between 1999 and 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide is the 10th leading cause of death and is one of just three leading causes that are increasing Additionally, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased 42% between 2001 and 2015 (1). Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (I). The Nations! iraiegjv for Suicide Preveniion (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyirelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factOr, but rather, is determined by multiple factors. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems, and preventing reattempts (6). In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic problems, access to lethal means substances, ?rearms) among persons at risk, and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the nation?s goal of reducing suicide rates by 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates and assessed the multiple contributing factors to suicide; this report presents options for multi-level comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for persons aged 3:10 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death ceni?cate records {International Classi?cation of Diseases I Revision, underlying-cause-of death codes Y87.0, U03). Age-specific population estimates were obtained from US. Census Bureaui?National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999?2016 (1999?2001, 2002?2004; 2005?2007'; 2008?2018; 2011? 2013; and 2014?2016). Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics of persons aged 210 years who died by suicide, with and without known mental health problems, and the circumstances surrounding the suicides were compared in the 27 states 3 with complete data participating in National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certi?cates, coronen?medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known mental health problems were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% con?dence intervals (CI), controlling for age group, sex, and race! ethnicity. RESULTS The most recent overall suicide rates (representing 2014?2016) varied four-fold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Supplementary Table). Across the study period, rates increased in all states except Nevada (where the rate was consistently high throughout the study period), with absolute increases ranging from +0.8 per 100,000 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +53% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (Supplementary Table) (Figure 1). Modeled suicide rate trends indicated signi?cant increases in 44 states, among males (34 states) and females (43 states), as well as for the United States overall (Supplementary Table). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated signi?cant increases for males (AAPC +1 . and females (AAPC (Supplementary Table). Suicide decedents without known mental health problems (N 11,039) were compared with those with known mental health problems (N 9,407) in 27 states. Whereas all decedents were predominately male (768%) (Table 1) and non?Hispanic white those without known mental health problems, relative to those with mental health problems, were more likely male (83.6% versus 68.8%; odds ratio (OR) 2.3, 95% CI 2.2?2.5) and racial/ethnic minorities (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide?suicide (adjusted odds ratio 2.9, 95% CI 2.2? 3.8). Among adult decedents aged 3:18 years, 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever served, or were currently serving, in the U.S. military. Whereas ?rearms were the most common method of suicide overall and among decedents with and without mental health problems, decedents without known mental health problems, relative to those with known mental health problems, were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% versus 31.3%) or poisoning (10.4% versus These differences remained signi?cant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall 0.8, 95% CI including opioids 0.90 95% CI but were more likely to test positive for alcohol 1.2, 95% CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (N 9,407) and approximately 85% of those without known mental health problems (N 9,357) in 27 states (Table 2). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% CI 0.7-0.8) than were persons with known mental health problems. Whereas two thirds of decedents with known mental health problems had a history of mental health or substance use treatment just over half were in current treatment. Decedents without known mental health problems had a significantly higher likelihood of any relationship problemfloss than did those with known mental health problems speci?cally intimate partner problems (30.2% versus (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than were those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminal-legal problems (10.7% versus or evictioni?loss of home versus and were more likely to have had a recent or impending (within the preceding or upcoming 2 weeks, respectively) crisis (a current or acute event thought to contribute to the suicide) (32.9% versus All of these differences remained signi?cant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobi??nancial problems were commonly experienced among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.8%, respectively). Decedents without known mental health problems had significantly lower odds of recent release from any institution 0.5, 95% CI but those who were recently released were significantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcohol/substance treatment) 2.5 95% CI than were those with a known mental health problem. Among decedents with known mental health problems who were recently released from an institution 46.7% of were released from facilities. Decedents without known mental health problems were significantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health problems, respectively. Conclusions and Comments During 1999?2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increases of more than 30%. Rates increased signi?cantly among males in 34 states, and females, in 43 states. This ?nding is consistent with prior research showing a decreasing gender gap in male-female suicide rates during 1999?2014 Additional research into the specific causes of these trends is necessary. Data from the 27 states participating in provides important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health problems and prevention of reattempts. This study found that more than half of suicide decedents in did not have a known mental health problems, indicating that additional focus on non-mental health factors, ?irther upstream, is essential to a public health approach (I0). This group suffered more from relationship problems and other life stressors such as criminal-legal matters, evictionz?loss of home, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as jobf?nancial andfor physical health problems. These ?ndings point to the need to both help persons manage the conditions associated with mental health problems in the ?rst place, and to support persons with known mental health problems to decrease their risk of poor outcomes Two thirds of this group had a history of any mental health andfor substance use treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive- behavioral therapies. Additionally, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (12). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household financial support); teaching coping and problem- solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicali?mental health problems) (12). Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention The ?ndings in this report are subject to at least three limitations. In the state-level analysis. rankings for four states (Maryland. Massachusetts. Rhode Island. and Utah) might have been affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second. is not yet nationally representative; the 27 states included represent 49.6% of the population 1. Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of infonnant knowledge can affect data completeness and accuracy. Studies including more in-depth interviews with next-of-kin often identify greater attributions to mental disorders however. many methodological variations across studies exist (M). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown. and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package Programs, and PraetieesUB) to better understand their suicide problenL prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard, and Margaret Warner, Division of Vital Statistics, National Center for Health Statistics, CDC. Con?ict of Interest No con?icts of interest were reported. 'Division of Violence Prevention, National Center for ury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488-3942 References 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. lvey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnovv-Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacelEthnicity, Age GrOup, and Mechanism of Death?United States, 2001?2015. MMWR Surveill 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 10 6. Zalsman G, Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 8. Crepeau-Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 9. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. 10. Caine ED, Reed Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017:injuryprev-2017-042366. 11. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. 12. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. 13. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc 14. Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav Summary What is already known about this topic? In 2016, nearly 45,000 lives were lost to suicide in the United States. 11 What is added by this report? During 1999?2016, suicide rates increased in nearly every state, including 330% increases in 25 states. Mental health problems otter: contribute to suicide; however, 2015 data from the National Violent Death Reporting System (27 states) indicate that 54% of suicide decedents were not known to have such problems. Other contributors included relationship, substance use, health, and jobf?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates by 2025. 12 5.3.18 Vital Signs: Trends in State Suicide Rates United States, 1999?2016 and Circumstances Contributing to Suicide 27 States, 2015 Deborah M. Stone, Thomas R. Simon Katherine A. Fowler, Scott R. Kegler, Pth; Keming Yuan, Kristin M. Holland, Asha Z. Ivey?Stephenson, Alex E. Crosby, Background: Suicide rates in the United States have risen nearly 30% since 1999, and mental health problems are just one factor contributing to suicide. Examining state-level trends in, and the multiple circumstances contributing to, suicide can inform comprehensive state suicide prevention planning. Methods: Trends in age?adjusted suicide rates among persons aged 210 years, by state and sex, across six consecutive three-year periods (1999?2016), were assessed using data from the National Vital Statistics System for 50 states and the District of Columbia (DC). Data from the National Violent Death Reporting System, covering 27 states in 2015, were used to examine contributing circumstances among decedents with and without known mental health problems. Results: During 1999?2016, suicide rates increased signi?cantly in 44 states, with 25 states experiencing increases of more than 30%. Rates increased signi?cantly among males and females in 34 and 43 states, respectively. In 2015, more than half of decedents in 27 states did not have a known mental health problem. Among persons with circumstance information, several circumstances were signi?cantly more likely among those without a known mental health problems than among decedents with mental health problems, including relationship problemsfloss [45.1% vs life stressors {54.2% vs and recentfimpending crises (32.9% vs but these circumstances were common across groups. 5.3.18 Conclusions: Suicide rates increased signi?cantly across most states during 1999?2016. Various circumstances contributed to suicides among persons with and without known mental health problems. Implications for Public Health Practice: States can use a comprehensive evidence?based public health approach to prevent suicide risk before it occurs, identify and support persons at risk, prevent reattempts, and help friendsifamily after a suicide occurs. INTRODUCTION BACKGROUND AND PURPOSE In 2016, nearly 45,000 suicides (15.6! 100,000 population [age?adjusted]) occurred in the United States, among persons aged 310 years (1). Between 1999 and 2015, suicide rates increased among both sexes, all racialfethnic groups, and all urbanization levels Suicide is the 10th leading cause of death and is one of just three leading causes that are increasing Additionally, rates of emergency department visits for nonfatal self-harm, a key risk factor for suicide, increased nearly 45% between 2001 and 2015 Together, suicides and self-harm injuries cost the nation more than $69 billion in direct medical and work loss costs (1). The Nations! iraiegjv for Suicide Preveniion (NSSP) (5) calls for a public health approach to suicide prevention with efforts spanning multiple levels individual, familyirelationship, community, and societal). Such a comprehensive approach underscores that suicide is rarely caused by any single factOr, but rather, is determined by multiple factors. Despite the NSSP guidance, suicide prevention largely focuses on identifying suicidal persons, providing treatment for mental health problems and preventing reattempts (6). In addition to mental health problems and prior attempts, other circumstances contributing to suicide include social and economic 5.3.18 problems, access to lethal means substances, ?rearms) among persons at risk, and poor coping and problem-solving skills (5). Expanded awareness of these additional circumstances contributing to suicide risk and action to address them can help reach the nation?s goal of reducing suicide rates 20% by 2025 (7). To assist states in achieving this goal, CDC analyzed state-speci?c trends in suicide rates, assessed the multiple contributing factors to suicide, and presents options for multi-level comprehensive suicide prevention based on the best available evidence. METHODS Suicide rates were analyzed for persons aged 3:10 years only, as determining suicidal intent in younger children can be dif?cult (8). Age-speci?c suicide counts were tabulated based on National Vital Statistics System coded death ceni?cate records {International Classi?cation of Diseases 10? Revision, underlying-cause-of death codes Y87.0, U03). Age-specific population estimates were obtained from US. Census Bureaui?National Center for Health Statistics bridged-race population data releases. National and state-level suicide rate estimates were calculated for six consecutive three-year aggregate periods spanning 1999?2016 (1999?2001, 2002?2004; 2005?2002; 2008?2018; 2011? 2013; and 2014?2016). Rate estimates were age-adjusted to the US. year 2000 standard population and expressed per 100,000 persons per year. Age-adjusted suicide rate trends were modeled using the same three-year data aggregates, employing weighted least squares regression with inverse-variance weighting. Modeled rate trends are reported in terms of average annual percentage changes (AAPCs). Characteristics and circumstances of persons aged 3:10 years who died by suicide, with and without known mental health problems, were compared in the 27 states with complete data 3 5.3.18 participating in CDC ?5 National Violent Death Reporting System in 2015. de?nes mental health problems as disorders and listed in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (9), with the exception of alcohol and other substance use disorders, which are captured separately in aggregates data from three primary data sources: death certi?cates, coroner/medical examiner reports (including toxicology), and law enforcement reports. Decedents with and without known mental health problems were compared using Chi-square tests. Logistic regression analyses estimated adjusted odds ratios with 95% con?dence intervals (CI), controlling for age group, sex, and raceiethnicity. RESULTS The most recent overall suicide rates (representing 2014?2016) varied four-fold, from 6.9 (DC) to 29.2 (Montana) per 100,000 persons per year (Table 1). Across the study period, rates increased in all states except Nevada (which had a consistently high rate throughout), with absolute increases ranging from +0.8 per 100,000 (Delaware) to +8.1 (Wyoming). Percentage increases in rates ranged from +59% (Delaware) to +57.6% (North Dakota), with increases of more than 30% observed in 25 states (Table 1) (Figure). Modeled suicide rate trends indicated signi?cant increases in 44 states (Table 1), among males (34 states) and females (43 states; Supplemental Table I), as well as for the United States overall (Table 1). Nationally, the model-estimated AAPC for the overall suicide rate was By sex, estimated national rate trends further indicated significant increases for males (AAPC and females (AAPC (Supplemental Table Suicide decedents without known mental health problems (N 11,039) were compared with those with known mental health problems (N 9,407) in 27 states. Whereas all decedents were predominately male (Table 2) and non-Hispanic white those without known 4 5.3.18 mental health problems, relative to those with mental health problems, were more likely male (83.6% versus 68.8%; odds ratio (OR) 2.3, 95% CI 2.2-2.5) and racialfethnic minorities (OR range: Suicide decedents without known mental health problems also had signi?cantly higher odds of perpetrating homicide-suicide (adjusted odds ratio 2.9, 95% C1 3.8). Among adult decedents 1318 years, 20.1% of those without known mental health problems and 15.3% of those with mental health problems had ever served, or were currently serving, in the U.S. military. Whereas ?rearms were the most common method of suicide overall and for decedents with and without mental health problems, decedents without known mental health problems were more likely to die by ?rearm (55.3% versus 40.6%) and less likely to die by (26.9% ver3us 31.3%) or poisoning (10.4% verSus 19.8%) than were those with known mental health problems. These differences remained signi?cant in the adjusted models. Decedents without known mental health problems were less likely to receive toxicology testing. Among those with toxicology results, decedents without known mental health problems were less likely to test positive for any substance overall 0.8, 95% C1 including opioids 0.90 95% CI but were more likely to test positive for alcohol 1.2, 95% CI Information on circumstances surrounding suicide were available for all decedents with mental health problems (N 9,407) and approximately 85% of those without known mental health problems (N 9,357) (Table 3). Persons without known mental health problems were less likely to have any substance use disorders 0.7, 95% Cl Whereas two thirds of 5.3.18 decedents with known mental health problems had a history of mental health or substance use treatment just over half were in current treatment. Decedents without known mental health problems had a signi?cantly higher likelihood of any relationship problemfloss than did those with known mental health problems specifically intimate partner problems (30.2% versus (17.5% versus and recently perpetrating interpersonal violence versus Decedents without known mental health problems were also more likely than those with known mental health problems to have experienced any life stressors (54.2% versus 49.7%) such as criminal-legal problems (10.7% versus or evictionx?loss of home versus and were more likely to have had a crisis a current or acute event thought to contribute to the suicide, within the preceding or impending, two weeks (32.9% versus All of these differences remained significant in the adjusted models. Among all persons with recent crises, intimate partner problems were the most common types and did not differ by group. Similarly, physical health problems and jobx??nancial problems were commonly experienced among both persons without mental health problems (23.2% and 15.6%, respectively) and those with mental health problems (21.4% and 16.8%, respectively). Decedents without known mental health problems had significantly lower odds of recent release from any institution 0.5, 95% CI but those who were recently released were signi?cantly more likely to have been released from a correctional facility (25.7% versus hospital (43.7% versus or other facility alcoholisubstance treatment) 2.5 95% CI than those with a known mental health problems. Among decedents with known mental health problems who were recently released from an institution 46.7% of were released from facilities. 5.3.18 Decedents without known mental health problems were signi?cantly less likely to have a history of suicidal ideation or prior suicide attempts compared with those with known mental health problems (40.8% and 29.4%, respectively). Suicide intent was disclosed by 22.4% and 24.5% of persons without and with known mental health problems, respectively. Conclusions and Comments During 1999-2016, suicide rates increased signi?cantly in 44 states, and 25 states experienced increases of more than 30%. Rates increased signi?cantly among males in 34 states, and females, in 43 states. This ?nding is consistent with prior research showing a decreasing gender gap in male?female suicide rates during 1999?2014 Additional research into the speci?c causes of these trends is necessary. Fortunately, data from the 27 states participating in provides important insight into circumstances surrounding suicide and can help states identify prevention priorities. Suicidologists regularly state that suicide is not caused by a single factor (5), however, suicide prevention is often oriented toward identi?cation of suicidal persons, treatment of mental health problems and prevention of reattempts. This study found that more than half of suicide decedents in did not have a known mental health problems, indicating that additional focus on non-mental health factors, further upstream, is essential to a public health approach (10). This group suffered more from relationship problems and other life stressors such as criminal-legal matters, evictionfloss of home, and recent or impending crises. Similarly, persons with mental health problems often experienced relationship problems and other life stressors such as jobr??nancial andfor physical health problems. These ?ndings point to the need to both help persons manage the conditions associated with mental health problems in the ?rst place, and to support persons with known mental health problems to decrease their risk 7 5.3.18 of poor outcomes Two thirds of this group had a history of any mental health andfor substance use treatment, with over half in treatment when they died. This suggests the need for additional safety supports, including broader implementation of affordable and effective treatment modalities such as doctor-patient collaborative care models and proven cognitive- behavioral therapies. Additionally, increased access to behavioral health providers in underserved areas is needed, as is expansion of health care systems that integrate physical and behavioral health, with a priority on suicide prevention and patient safety, especially through care transitions (I2). Comprehensive statewide suicide prevention activities are needed to address the full range of factors contributing to suicide. Prevention strategies include strengthening economic supports housing stabilization policies, household ?nancial support); teaching coping and problem- solving skills to manage everyday stressors and prevent future relationship problems, especially early in life; promoting social connectedness to increase a sense of belonging and access to informational, tangible, emotional, and social support; and identifying and better supporting persons at risk military veterans, persons with physicalx?mental health problems) (I2). Other strategies include creating protective environments reducing access to lethal means among persons at risk, creating organizational and workplace policies to promote help-seeking, easing transitions into and out of work for persons with mental health problems and other life challenges), supporting family and friends after a suicide, and assuring safe reporting by the media in order to prevent suicide contagion (i 2). Some states, such as Colorado, are planning to implement such a comprehensive approach to suicide prevention (10). The ?ndings in this report are subject to at least three limitations. In the state-level analysis, rankings for four states (Maryland, Massachusetts, Rhode Island, and Utah) might have been 5.3.13 affected by large proportions of injury deaths of undetermined intent (potentially biasing reported suicide rates downward), or decreased percentages of such deaths over time (potentially biasing estimated rate trends upward). Second, is not yet nationally representative; the 27 states included represent 49.6% of the population Finally, abstractors of data are limited to information contained in investigative reports. Therefore, the extent of infonnant knowledge can affect data completeness and accuracy. Studies including more in?depth interviews with next?of?kin often identify greater attributions to mental disorders however, many methodological variations across studies exist (14). It is likely that some persons without known mental health problems in the current study were experiencing mental health challenges that were unknown, and hence underreported by key informants. Nonetheless, the high prevalence of diverse contributing circumstances among those with and without known mental health problems suggests the importance of addressing the broad range of factors that contribute to suicide. Suicide is a growing public health problem. Effective approaches to prevent the many suicide risk factors are available. States and communities can use data from and resources such as Preventing Suicide: a Technical Package Programs, and Practices {12) to better understand their suicide problem, prioritize evidence-based comprehensive suicide prevention, and save lives. Acknowledgments Robert Anderson, Holly Hedegaard, and Margaret Wamer, Division of Vital Statistics, National Center for Health Statistics, CDC. Conflict of Interest No con?icts of interest were reported. 9 5.3.18 'Division of Violence Prevention, National Center for .Injury Prevention and Control, 2Division of Analysis, Research, and Practice Integration, National Center for Injury Prevention and Control, CDC. Corresponding author: Deborah M. Stone, dstone3@cdc.gov 770-488?3942 References 1. Centers for Disease Control and Prevention. Web-based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Retrieved March 15, 2018. In. 2. Ivey-Stephenson AZ, Crosby AE, Jack SPD, Haileyesus T, Kresnow?Sedacca MJ. Suicide Trends Among and Within Urbanization Levels by Sex, RacefEthnicity, Age Group, and Mechanism of Death?United States, 2001?2015. MMWR Surveill Summ 3. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. In: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics Hyattsville, 2016. 4. Kochanek K, Murphy S, Xu J, Arias E. Mortality in the United States, 2016. NCHS Data Brief, no 293. National Center for Health Statistics 2017. 5. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and objectives for action: a report of the US Surgeon General and of the National Action Alliance for Suicide Prevention. 2012. 6. Zalsman G, Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10-year systematic review. Lancet 7. Torguson K, O'Brien A. Leading Suicide Prevention Efforts Unite to Address Rising National Suicide Rate. In. Washington, 2017. 10 5.3.18 10. 11. 12. 13. 14. Crepeau?Hobson F. The autopsy and determination of child suicides: a survey of medical examiners. Arch Suicide Res 2010; 14:24n34. American Association. Diagnostic and statistical manual of mental disorders American Pub; 2013. Caine ED, Reed], Hindman J, Quinlan K. Comprehensive, integrated approaches to suicide prevention: practical guidance. Inj Prev 2017:injuryprev-2017-042366. World Health Organization. Risks to mental health: An overview of vulnerabilities and risk factors. Geneva: WHO 2012. Stone DM, Holland KM, Bartholow BN, Crosby AE, Davis SP, Wilkins N. Preventing suicide: A technical package of policies, programs, and practice. 2017. Milner A, Sveticic J, De Leo D. Suicide in the absence of mental disorder? A review of autopsy studies across countries. Int Soc Pouliot L, De Leo D. Critical issues in autopsy studies. Suicide Life Threat Behav 10. Summary Box (word count 1001100) What is already known on the topic? In 2016, nearly 45,000 lives were lost to suicide in the US. What is added by this report? Between 1999-2016, suicide rates increased in nearly every state. Twenty?five states saw rate increases Mental health problems often contribute to suicide, however, 2015 data from the National Violent Death Reporting System (27' states) indicate that 54% of suicide decedents 11 5.3.18 were not known to have such problems. Other contributors included relationship, substance use, health, and jobf?nancial problems. What are the implications for public health practice? A comprehensive approach using proven prevention strategies, such as those in Technical Package for Suicide Prevention, can help reach the national goal of reducing suicide rates 20% by 2025. 12 Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 109,999 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2091 2092 2094 2095 2997 2906 2010 2011 2913 2914 2016 Rank (State Rank) '9 (State Ragnk) H. Both 12.3 (We) 12.7 9.4) 12.9 9.2) 13.6 0.9) 14.5 9.6) 15.4 9.9) 1.5 ?70 (pr-1.91) rda 3.1 (nia) 25.4 (nia) U.S. Male 29.9 (nia) 21.2 9.4) 21.3 9.9) 22.5 23.5 1.9) 24.5 1.9) Female 4.7 5.9 9.3) 5.3 9.2) 5.7 0.4) 6.2 9.5) 6.9 9.7) 2.6 ?is Both 14.3 (nia) 13.4 (- 9.9) 14.1 9.6) 15.6 1.6) 16.4 9.7) 17.5 .1) +1.6 (ps. 95) 25 3.1 (31) 21.9 (33) AL Male 25.1 (nia) 23.4 (- 1.7) 24.4 1.0) 26.4 2.0) 27.6 1.1) 29.1 1. 5) 1. 3 (be. 95) Female 5.1 (nia) 4.6 (- 0.3) 5.0 0.2) 1 6.4 9.3) 9. 7) 2. 6 (ps. 01) Both 21.9 (his) 24.6 3.6) 24.2 (- 9.6) 26.9 1.7) 25.4 (- 9.5) 26.6 3.4) +1 7 (p4. 95) 2 7.6 4) 37.4 (13) AK Male 33.2 (nia) 36.1 4.9) 36.9 9.6) 49.1 49.1 (- 0.1) 42. 9 2. 6) +1.4 (ps. 91) Female 66 (nia) 11.4 2.9) 11.1 9.9 (- 1.his Both 17.6 (nia) 16.5 9.7) 19.1 0.5) 19.1(- 9.9) 29. 4 1.3) 29.9 9. 5) +1.9 91) 15 3.1 (32) 17.3 (42) AZ Male 29.3 (nia) 30.2 1.9) 39. 6 0.4) 39.2 0.5) 32. 0 1.(ps. 95) Female 7.1 (nia) 7.5 9.4) 9.7) 0.5) 9.6) 9. 6) 2. 2 (ps. 91) Both 15.5 (his) 15.6 9.3) 16.2 9.5) 17.6(+ 1.4) 19.2 1.6) 21 91) 12 +5.7(14) AR Male 26 7 (nia) 26.7 0.9) 27. 2 9.5) 26. 2 1.0) 31. 7 3.(ps. 05) Female 5.6 (nia) 5.9 9.3) 6. 2 9 4) 1.7) 7.5 (- 0.4) 2.1) 3. 6 (.ps 91) Both 19.6 (his) 11.3 11 03) 11.6 (- 9.1) 121 05) 45 1.6 (46) CA Male 17.9 (nia) 16.4(+ 9.5) 17.7-( 9.7) 19.1 16.9 (-0.2) +95% his Female 4.1 (nia) 5.9 0.9) 4.9 91) 9.5) 5. 3 (- 9.1) 9. 3) 7 (ps. 05) Both 17.3 (his) 19.2 1.9) 19. 9 .2) 29. 9 1.9) 21.6 1.5) 23.2 16) 1.6 (p4. 91) 6 5.9 (12) 34.1 (22) CO Male 26.6 (nia) 39.9 2.3) 39.5 9 .4) 31.5 1.0) 33.4 1.9) 36. 3 2. 9) +1.4 (ps. 91) Female 7.0 (nia) 6.2 1.3) 6.2 0.0) 0.(be. 91) Both 9.6 (nia) 6.9 9.7) 9.1 0.2) 19. 2 1.1) 11.9 0.6) 11.5 9.5) +1.6 (be. 95) 46 1.9 (43) +192 (34) CT Male 16.4 (nia) 14.6 (w 1.6) 15.0 9.4) 16 6 1.6) 17. 6 1.9) 17.3 (- 9.3) 9.9 his Female 3.6 (nia) 3.6 9.2) 3.7 (- 0.2) 0.7) 9.5) 6.2 1.3) 3.5 Rates are age-adjusted to the US. year 2999 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2914 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase to largest percentage decrease (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 29 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 1? (State ank) Both 13.6 (nia) 12. 211 .4) 11.91; 0.3) 13.61+ 1.7) 14.21+ 0.6) 14.41+ 0.2) 0.9 42 0.8150) 5.9 311150) DE Male 23. 0 (Na) 20. 31-2 .7) 19.91- 0.4) 23.1 1+ 3.2) 22. 7 1- 0. 4) 2351+ 0.8) 0.6 ?fa nis Female 31nla) 010.2) 4.614 0.4) 91+ 0.3) 6. 41 1 5.) 21- 0.2) 1.6 his Both 91nla) 641+ 0.5) 6.41? 0.0) 31+ 0.6) 6?01.7) 6.91+ 0.3) 0.9 36 nis 51 1.0146) 16.1 ?is 145) DC Male 10.7 (nla) 11.11+ 0.4) 10.31; 0.8) 1271+ 2.4) 10. 012 .6) 11. 71+ 1.7) 0.3 ?it: nis Female 1.7 i1riia)?FT 231+ 0.6) TT 3.31+ 1.0) 61- 0.7) 3.61+ 1.0) 810.8) 3.5 ?fa nis Both 14.8 (nia) 1521+ 0.4) 14.91- 0.3) 16. 3 1+ 1.4) 1631.0) 16.41+ 0.1) 0.8 361p<.05) 29 1.6145) 10.6 i13146) FL Male 24.3 (We) 24.41+ 0.1) 23.61- 0.8) 26. 2 1+ 2.6) 25. 61 0. 6) 25.6 1) 0.5 36 Female 6.3 (nia) 6.61+ 0.5) 681+ 0.0) 11+ 0.3) 7.61+ 0.5) 81+ 0.3) 1.4 901p<.01) Both 12.91nla) 13.21+ 0.3) 12.31? 0.9) 13. 2 1+ 0.9) 13.71+ 0.5) 1501+ 1.3) 0.9 ?it; nis 39 2.1 140) +162 13144} GA Male 22.1 (his) 23.1 1+ 1 0.) 21.3 1? 1.8) .91+ 0.6) 2261+ 0.7) 24.-41+ 1.7) 0.5 ?fa his Female 5.0 (Na) 810.2) 4.6 1- 0.2) 51+ 0.9) 5.8 1+ 0.3) 6.6 1+ 0.8) 2.1 ?fa 1p<.05) Both 12.91nla) 11. 1 11.6) 10.31- 0.7) 14.51+ 4.1) 14.41? 0.1) 1521+ 0.8) 2.0 ?fa nis 35 2.4135) +183 I1111136) HI Male 20.4 (Ma) 17. 21-3 .1) 15.3 1- 1.9) 21. 91+ 6.7) 2251+ 0.5) 24.31+ 1.8) 2.1 ?fa nis Female 5.4 (We) 010.4) 5.51+ 0.5) 11+ 1.5) 21- 0 9) 5.91- 0.3) 1.2 ?fa his Both 17.3 1111a) 1921+ 2.0) 16.31? 0.9) 21. 61+ 3.3) 2191+ 0.3) 2471+ 2.8) 2.3 3611:1101) 6 7.51 6) 43.2 1 7) ID Male 28.4 (nia) 33.1 1+ 4.7) 31.1 1? 2.0) 3 .91+ 3.8) 71- 0 2) 3801+ 3.3) 1.6 301p<.05) Female 7.2 (Na) 6.1 1- 1.1) 6.1 1+ 0.0) 01+ 2.9) 9.51+ 0.5) 1181+ 2.3) 4.4 1p<.05) Both 9.9 (nia) 9.81? 0.1) 9.71- 0.1) 10. 6 1+ 0.8) 11.2 1+ 0.6) 12.21+ 1.0) 1.5 44 2.3138) 22.8 ?it (32) IL Male 17.1 (We) 16.71? 0.4) 16.2 1- 0.4) 17. 61+ 1.4) 1851+ 0.9) 1981+ 1.3) 1.1 3611:1105) Female 3.7 (nla) 3.814 0.0) 381+ 0.2) 21+ 0.4) 4.51+ 0.4) 521+ 0.6) 2.4 301p<.01) Both 13.01n1a) 13.71+ 0.7) 14. 4 1+ 0. 7) 14.91+ 0.5) 16.41+1.4) 17.11+ 0.7) +1.9 1p<.01) 26 4.1123) 31.9 13125} IN Male 22.4 (his) 2321+ 0.8) 24. 41+ 1 .2) 24. 71+ 0.4) 2671+ 2.0) 2831+ 1.6) 1.5 3611:1101) Female 4.6 (We) 501+ 0.4) 531+ 0.2) 91+ 0.6) 681+ 0.9) 6.61? 0.2) 2.7 3611:1101) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2016) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nfa) 22.1 1.5) 20.8 .4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nfa) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 i111(ps0?i) 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nfa) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nta) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nfa) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia) 12.9 0.2) 13. 4 0 .4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 31; 27 3.8 29.3 as (26) LA Male 22.9 (nfa) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa nis Female 4.8 (nfa) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 3?0 Both 14.5 (nia) 1310) 18.9 3.5) 18. 0.4) 2.2 31: 21 4.0 (25) 2?.4 ?11; (29) ME Male 25.0 (nfa) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nfa) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nfa) 10.3 0.3) 10.1 0.2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nfa) 3.8 0.4) 0.0) 7(0.2) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nfa) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 900 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (We) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 as (24) Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nfa) 4.8 0.0.9) 2.8 3?0 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. 1 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 11.-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 (We) 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 41n1a1 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 (Ma) 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.51n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20151 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51n7a) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30 Till) 34.1 ?fa (23 RI Male 15.4 (nfa) .2) 14.8(- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.16.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (nfa) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (nfa) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18.2 1.0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 26.8 1.3) 8 0 1. 2) 28.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (nfa) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (nfa) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3 1W) 46.5 3?a( 4 UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (nfa) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2(+4 0. 9) 6.4 1.3) 6 6 0.2) 7. 3 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (nfa) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 25.1 1+1 .0) 25. a 1+ 1:1. 9) 27.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nl?s indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons :2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons (Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2991 2992 2994 2995 2997 2998 2919 2911 2913 2914 2916 Rank (State Rank) 1i (State Ragnk) H. Both 12.3 (Ma) 12.7 9.4) 12.9 9.2) 13.8 9.9) 14.5 9.8) 15.4 9.9) 1.5 fit. (pr-1.91) nia 3.1 (nia) 25.4 (nia) U.S. Male 29.9 (nia) 21.2 9.4) 21.3 9.9) 22.5 23.5 1.9) 24.5 1.9) Female 4.7 5.9 9.3) 5.3 9.2) 5.7 9.4) 6.2 9.5) 6.9 9.7) 2.6 31. Both 14.3 (nia) 13.4 (- 9.9) 14.1 9.6) 15.6 1.6) 16.4 9.7) 17.5 .1) +1.6 (ps. 95) 25 3.1 (31) 21.9 (33) AL Male 25.1 (nia) 23.4 (- 1.7) 24.4 1.9) 26.4 2.9) 27.6 1.1) 29.1 1. 5) 1. 3 (ps. 95) Female 5.1 (nia) 4.8 (- 9.3) 5.9 9.2) 1 6.4 9.3) 9. 7) 2. 6 (.ps 91) Both 21.9 (nia) 24.8 3.8) 24.2 (- 9.6) 26.9 1.7) 25.4 (- 9.5) 28.8 3.4) +1 7 (p4. 95) 2 7.8 4) 37.4 (13) AK Male 33.2 (nia) 38.1 4.9) 38.9 9.8) 49.1 49.1 (- 9.1) 42. 9 2. 8) +1.4 (ps. 91) Female 8.6 (nia) 11.4 2.9) 9.8 (- 1.6) 11.1 9.9 (- 1.his Both 17.8 (nia) 18.5 9.7) 19.1 9.5) 19.1(- 9.9) 29.4 1.3) 29.9 9. 5) +1.9 (.ps 91) 15 3.1 (32) 17.3 (42) .42 Male 29.3 (nia) 39.2 1.9) 39. 6 9.4) 39. 2 9.5) 32. 9 1.(ps. 95) Female 7.1 (nia) 7.5 9.4) 9.7) 9.5) 9.6) 9. 6) 2. 2 (ps. 91) Both 15.5 (nia) 15.8 9.3) 16.2 9.5) 17.6(+ 1.4) 19.2 1.6) 21 91) 12 +5.7(14) AR Male 26 7 (nia) 26.7 9.9) 27. 2 9.5) 28. 2 1.9) 31. 7 3.5) 33. 5 1. 9) 1.6 (ps. 95) Female 5.6 (nia) 5.9 9.3) 6. 2 9 4) 1.7) 7.5 (- 9.4) 2.1) 3. 6 (.ps 91) Both 19.6 (nia) 11.3 11 93) 11.8 (- 9.1) 12.1 95) 45 1.6 (46) CA Male 17.9 (nia) 18.4 9.5) 17.7-( 9.7) 19.1 18.9 (-9.2) +95% his Female 4.1 (nia) 5.9 9.9) 4.9 9.1) 9.5) 5. 3 (- 9.1) 9. 3) 7 (ps. 95) Both 17.3 (nia) 19. 9 .2) 29. 9 1.9) 21.6 1.5) 23.2 16) 1.8 (p4. 91) 8 5.9 (12) 34.1 (22) CO Male 28.6 (nia) 39.9 2.3) 39.5 (9 .4) 31.5 1.9) 33. 4 1.9) 36. 3 2. 9) +1.4 (ps. 91) Female 7.9 (nia) 8.2 1.3) 8.2 9.(ps. 91) Both 9.6 (nia) 8.9 (w 9.7) 9.1 9.2) 19. 2 1.1) 11.9 9.8) 11.5 9.5) +1.6 (ps. 95) 46 1.9 (43) +192 (34) CT Male 16.4 (nia) 14.6 (w 1.8) 15.9 9.4) 16. 6 1.6) 17. 6 1.9) 17.3 (- 9.3) 9.9 his Female 3.6 (nia) 3.8 9.2) 3.7 (- 9.2) 9.7) 9.5) 6.2 1.3) 3.5 Rates are age-adjusted to the US. year 2999 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first and last periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. if Rate based on 29 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -1- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 1? (State ank) Both 13.6 (nia) 12. 211 .4) 11.91; 0.3) 13.61+ 1.7) 14.21+ 0.6) 14.41+ 0.2) 0.9 42 0.8150) 5.9 311150) DE Male 23. 0 (Na) 20. 31-2 .7) 19.91- 0.4) 23.1 1+ 3.2) 22. 7 1- 0. 4) 2351+ 0.8) 0.6 ?fa nis Female 31nla) 010.2) 4.614 0.4) 91+ 0.3) 6. 41 1 5.) 21- 0.2) 1.6 his Both 91nla) 641+ 0.5) 6.41? 0.0) 31+ 0.6) 6?01.7) 6.91+ 0.3) 0.9 36 nis 51 1.0146) 16.1 ?is 145) DC Male 10.7 (nla) 11.11+ 0.4) 10.31; 0.8) 1271+ 2.4) 10. 012 .6) 11. 71+ 1.7) 0.3 ?it: nis Female 1.7 i1riia)?FT 231+ 0.6) if 3.31+ 1.0) 61- 0.7) 3.61+ 1.0) 810.8) 3.5 ?fa nis Both 14.8 (nia) 1521+ 0.4) 14.91- 0.3) 16. 3 1+ 1.4) 1631.0) 16.41+ 0.1) 0.8 361p<.05) 29 1.6145) 10.6 i13146) FL Male 24.3 (We) 24.41+ 0.1) 23.61- 0.8) 26. 2 1+ 2.6) 25. 61 0. 6) 25.6 1) 0.5 36 nis Female 6.3 (nia) 6.61+ 0.5) 681+ 0.0) 11+ 0.3) 7.61+ 0.5) 81+ 0.3) 1.4 901p<.01) Both 12.91nla) 13.21+ 0.3) 12.31? 0.9) 13. 2 1+ 0.9) 13.71+ 0.5) 1501+ 1.3) 0.9 ?it; nis 39 2.1 140) +162 13144} GA Male 22.1 (his) 23.1 1+ 1 0.) 21.3 1? 1.8) .91+ 0.6) 2261+ 0.7) 24.-41+ 1.7) 0.5 ?fa his Female 5.0 (Na) 810.2) 4.6 1- 0.2) 51+ 0.9) 5.8 1+ 0.3) 6.6 1+ 0.8) 2.1 ?fa 1p<.05) Both 12.91nla) 11. 1 11.6) 10.31- 0.7) 14.51+ 4.1) 14.41? 0.1) 1521+ 0.8) 2.0 ?fa nis 35 2.4135) +183 I1111136) HI Male 20.4 (Ma) 17. 21-3 .1) 15.3 1- 1.9) 21. 91+ 6.7) 2251+ 0.5) 24.31+ 1.8) 2.1 ?fa nis Female 5.4 (We) 010.4) 5.51+ 0.5) 11+ 1.5) 21- 0 9) 5.91- 0.3) 1.2 ?fa his Both 17.3 (nia) 1921+ 2.0) 16.31? 0.9) 21. 61+ 3.3) 2191+ 0.3) 2471+ 2.8) 2.3 3611:1101) 6 7.51 6) 43.2 1 7) ID Male 28.4 (nia) 33.1 1+ 4.7) 31.1 1? 2.0) 3 .91+ 3.8) 71- 0 2) 3801+ 3.3) 1.6 301p<.05) Female 7.2 (Na) 6.1 1- 1.1) 6.1 1+ 0.0) 01+ 2.9) 9.51+ 0.5) 1181+ 2.3) 4.4 1p<.05) Both 9.9 (nia) 9.81? 0.1) 9.71- 0.1) 10. 6 1+ 0.8) 11.2 1+ 0.6) 12.21+ 1.0) 1.5 i5131p<.05) 44 2.3138) 22.8 ?it (32) IL Male 17.1 (We) 16.71? 0.4) 16.2 1- 0.4) 17. 61+ 1.4) 1851+ 0.9) 1981+ 1.3) 1.1 3611:1105) Female 3.7 (nla) 3.814 0.0) 381+ 0.2) 21+ 0.4) 4.51+ 0.4) 521+ 0.6) 2.4 301p<.01) Both 13.01n1a) 13.71+ 0.7) 14. 4 1+ 0. 7) 14.91+ 0.5) 16.41+1.4) 17.11+ 0.7) +1.9 1p<.01) 26 4.1123) 31.9 13125} IN Male 22.4 (his) 2321+ 0.8) 24. 41+ 1 .2) 24. 71+ 0.4) 2671+ 2.0) 2831+ 1.6) 1.5 3611:1101) Female 4.6 (We) 501+ 0.4) 531+ 0.2) 91+ 0.6) 681+ 0.9) 6.61? 0.2) 2.7 3611:1101) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; his indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -2- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 .4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nfa) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nta) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nfa) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia) 12.9 0.2) 13. 4 0 .4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 31; 27 3.8 29.3 as (26) LA Male 22.9 (nfa) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa nis Female 4.8 (nfa) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 3?0 Both 14.5 (nia) 1310) 18.9 3.5) 18. 0.4) 2.2 31: 21 4.0 (25) 2?.4 ?it; (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nfa) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nfa) 10.3 0.3) 10.1 0.2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(0.2) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nfa) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 900 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (We) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 as (24) Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nfa) 4.8 0.0.9) 2.8 3?0 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. 1 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -3- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 fit-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 1n1a1 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 4 (ma) 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 1n1a1 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.5 1n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -4- Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51nia) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -5- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30) 34.1 ?Va (23) RI Male 15.4 (Ma) .2) 14.8[- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa) 3-0( .7) 0. 4) 1. 3) 00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.2) 1316.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (Na) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (We) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18. 2 1 .0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 2628.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (his) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (Ma) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 611+ 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3) 46.5 ?if? 4) UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (We) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2H4 0. 9) 6.4 1.3) 6 6 0.2) 7. 31+ 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (We) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -5- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 2527.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Rani-ts are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1. Percentage Changes in Annual Suicide Rates (per 100,000, Age-Adjusted) 2014-2016 Compared Against 1999-2001 I: Decrease 1.0% I:Ilncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 316% - 57.8% Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age. by EtatJ?end-Sem-??wm state 59x Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period] hr: Egg-:31 Raguglrlil;ge 1999 - 2001 2002 2004 2005 - 2007 2008 2010 2011 - 2013 2014 2015 Rank {State Rank] 1 (?State Ragnk] Both 12.311113] 1271+ 0.4] 1291+ 0.2] 1351+ 0.9] 1451+ 0.5] 1541+ 0.9] 1.5 31: 1p<.01] ni'a 3.1 {rite} 25.4 %1n1ai US. Male 20.91nia] 2121+ 0.4] 21.31+ 0.0] 22.5 2351+ 1.0] 2451+ 1.0] 1.1 1p<.01] Female 4.71nia] 5.01+ 0.3] 531+ 0.2] 571+ 0.4] 521+ 0.5] 591+ 0.7] 2.5 1p<.01] Eloth 14.31nia] 13.41- 0.9] 1411+ 0.5] 1551+ 1.5] 15.41+ 0.7] 17.51+1.1] +1.5 ?its 1p<.05] 25 3.1 131] +219 913133} AL Male 25.1 1nia] 23.41-17] 24.41+1.0] 2541+ 2.0] 27.51+1.1] 29.1 1+ 1.5] 1.3 0A1p<.05} Female 5.1 1nia] 4.31- 0.3] 501+ 0.2] 5.1 1+ 1.1] 541+ 0.3] 701+ 0.7] 25 '31: Both 21.01nia] 2451+ 3.5] 24.21- 05] 2501+ 1.7] 25.41- 0.5] 2351+ 3.4] 1.7 ?fa 1p<.051 2 7-31 41 37.4 113] AK Male 33.21n1a] 35.1 1+ 4.9] 3591+ 0.5] 40.1 1+ 1.2] 40.1 1- 0.1] 4291+ 2.5] 1.4 ?i?a 1p<.01] Female 5.51nie] 1141+ 2.9] 9.51-1.51 11.1 1+ 1.2] 9.91- 1.2] 1321+ 3.4] 1.7 r115 Both 17.51n1a] 1551+ 0.7] 19.1 1+ 0.5] 19.1 1- 0.0] 20.41+1.3] 2091+ 0.5] 1.0 0111(pc.01} 15 3.1 132] +113 ?11: 142] AZ Male 29.31nia] 3021+ 1.0] 3051+ 0.4] 30.21? 0.5] 3201+ 1.9] 3241+ 0.4] 0.5 ?In 104.05] Female 7.1 1n1a] 7.51+ 0.4] 521+ 0.7] 551+ 0.5] 921+ 0.5] 991+ 0.5] 2.2 1p<.01] Both 15.5 (rite) 1531+ 03] 1521+ 0.5] 1751+ 1.4] 19.21+1.5] 2121+ 2.0] 2.2 12 5.7114] 35.5 ?xi: 115] AR Male 25.?1nia] 25.? 1+ 0.0] 2721+ 0.5] 3171+ 3.5] 3351+ 1.9] 1.5 ?it: 1p<.05] Female 5.51nia] 591+ 521+ 0.4] 791+ 1.7] 7.51- 0.4] 951+ 2.1] 3.5 991p{.01] Both 10.51nia] 1131+ 0.7] 11.01- 0.3] 1201+ 1.0] 11.31- 0.1] 12.1 1+ 0.3] 0.9 0:1: 1p<.05] 45 +1.5145] 14.3 145] CA Male 17.91n1a] 1341+ 0.5] 1171+ 0.7] 19.1 1+ 1.4] 15.912 0.2] 1921+ 0.3] 0.5 1115 Female 4.1 We] 501+ 0.9] 4.91? 0.1] 541+ 0.5] 5.31? 0.1] 551+ 0.3] 1.7 Both 17.31n1a) 1921+ 1.9] 19.01- 0.2] 2001+ 1.0] 21.51+1.5] 2321+ 1.5} 1.3 ?it: 1p<.01] 5 5.9112] 34.1 31: 122] CD Male 25.51n1a] 3091+ 2.3] 30.51- 0.4] 3151+ 1.0] 33.41+1.9] 3531+ 2.9} 1.4 925113?5111] Female 7.011113] 321+ 1.3] 821+ 0.0] 9.1 1+ 0.9] 10.1 1+ 1.0] 1041+ 0.3] 2.5 113101] Both 9.51nia] 5.91? 0.7] 9.1 1+ 0.2] 10.21+ 1.1] 1101+ 0.5] 1151+ 0.5] 1.5 ?fu1ps.05] 45 +1.9143] 19.2 ?it: 134] CT Male 15.41n1a] 1451-15] 1501+ 0.4] 15.51+ 1.5] 17.51+1.0] 17.31- 0.3] 0.9 n15 Female 3.51nl?a] 351+ 0.2] 3.71- 0.2] 4.41+ 0.7] 491+ 0.5] 521+ 1.3} 3.5 %1p<.05] decrease 151]. Differences between ranks do not necessarily imply a statistically signi?cant difference. 1'1 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the ?rst and last periods and might have contributed to lower reported rates. Rates are age-adjusted to the U5. year 2000 standard. Modelaestimaled average annual percentage change based on all reporting periods; p-value indicates statistical signi?cance of trend; nis indicates trend not significant 15 Current state rank 150 states and the District of Columbia] is forthe reporting period 2014 - 2015. Ranks are from highest rate 11] to lowest rate 151]. Differences between ranks do not necessarily imply a statistically signi?cant difference. Dyerell rate change is between the ?rst 11999 2001 i and Iast12014 2015] reporting periods. Ranks are from largest increase 11] to largest decrease 151]. Differences bent-teen ranks do not necessarily imply a statisticall signi?cant differenceComment Wonder ifin a revised title or footnote ifit needs to be revised to reflect this also includes District of Columbia. Comment Suggest being clear on what these period, similar to how done In another footnote. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 1? (State ank) Both 13.6 (nia) 12. 211 .4) 11.91; 0.3) 13.61+ 1.7) 14.21+ 0.6) 14.41+ 0.2) 0.9 42 0.8150) 5.9 311150) DE Male 23. 0 (Na) 20. 31-2 .7) 19.91- 0.4) 23.1 1+ 3.2) 22. 7 1- 0. 4) 2351+ 0.8) 0.6 ?fa nis Female 31nla) 010.2) 4.614 0.4) 91+ 0.3) 6. 41 1 5.) 21- 0.2) 1.6 his Both 91nla) 641+ 0.5) 6.41? 0.0) 31+ 0.6) 6?01.7) 6.91+ 0.3) 0.9 36 nis 51 1.0146) 16.1 ?is 145) DC Male 10.7 (nla) 11.11+ 0.4) 10.31; 0.8) 1271+ 2.4) 10. 012 .6) 11. 71+ 1.7) 0.3 ?it: nis Female 1.7 i1riia)?FT 231+ 0.6) if 3.31+ 1.0) 61- 0.7) 3.61+ 1.0) 810.8) 3.5 ?fa nis Both 14.8 (nia) 1521+ 0.4) 14.91- 0.3) 16. 3 1+ 1.4) 1631.0) 16.41+ 0.1) 0.8 361p<.05) 29 1.6145) 10.6 i13146) FL Male 24.3 (We) 24.41+ 0.1) 23.61- 0.8) 26. 2 1+ 2.6) 25. 61 0. 6) 25.6 1) 0.5 36 nis Female 6.3 (nia) 6.61+ 0.5) 681+ 0.0) 11+ 0.3) 7.61+ 0.5) 81+ 0.3) 1.4 901p<.01) Both 12.91nla) 13.21+ 0.3) 12.31? 0.9) 13. 2 1+ 0.9) 13.71+ 0.5) 1501+ 1.3) 0.9 ?it; nis 39 2.1 140) +162 13144} GA Male 22.1 (his) 23.1 1+ 1 0.) 21.3 1? 1.8) .91+ 0.6) 2261+ 0.7) 24.-41+ 1.7) 0.5 ?fa his Female 5.0 (Na) 810.2) 4.6 1- 0.2) 51+ 0.9) 5.8 1+ 0.3) 6.6 1+ 0.8) 2.1 ?fa 1p<.05) Both 12.91nla) 11. 1 11.6) 10.31- 0.7) 14.51+ 4.1) 14.41? 0.1) 1521+ 0.8) 2.0 ?fa nis 35 2.4135) +183 I1111136) HI Male 20.4 (Ma) 17. 21-3 .1) 15.3 1- 1.9) 21. 91+ 6.7) 2251+ 0.5) 24.31+ 1.8) 2.1 ?fa nis Female 5.4 (We) 010.4) 5.51+ 0.5) 11+ 1.5) 21- 0 9) 5.91- 0.3) 1.2 ?fa his Both 17.3 (nia) 1921+ 2.0) 16.31? 0.9) 21. 61+ 3.3) 2191+ 0.3) 2471+ 2.8) 2.3 3611:1101) 6 7.51 6) 43.2 1 7) ID Male 28.4 (nia) 33.1 1+ 4.7) 31.1 1? 2.0) 3 .91+ 3.8) 71- 0 2) 3801+ 3.3) 1.6 301p<.05) Female 7.2 (Na) 6.1 1- 1.1) 6.1 1+ 0.0) 01+ 2.9) 9.51+ 0.5) 1181+ 2.3) 4.4 1p<.05) Both 9.9 (nia) 9.81? 0.1) 9.71- 0.1) 10. 6 1+ 0.8) 11.2 1+ 0.6) 12.21+ 1.0) 1.5 i5131p<.05) 44 2.3138) 22.8 ?it (32) IL Male 17.1 (We) 16.71? 0.4) 16.2 1- 0.4) 17. 61+ 1.4) 1851+ 0.9) 1981+ 1.3) 1.1 3611:1105) Female 3.7 (nla) 3.814 0.0) 381+ 0.2) 21+ 0.4) 4.51+ 0.4) 521+ 0.6) 2.4 301p<.01) Both 13.01n1a) 13.71+ 0.7) 14. 4 1+ 0. 7) 14.91+ 0.5) 16.41+1.4) 17.11+ 0.7) +1.9 1p<.01) 26 4.1123) 31.9 13125} IN Male 22.4 (his) 2321+ 0.8) 24. 41+ 1 .2) 24. 71+ 0.4) 2671+ 2.0) 2831+ 1.6) 1.5 3611:1101) Female 4.6 (We) 501+ 0.4) 531+ 0.2) 91+ 0.6) 681+ 0.9) 6.61? 0.2) 2.7 3611:1101) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; his indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -2- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 .4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nfa) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nta) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nfa) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia) 12.9 0.2) 13. 4 0 .4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 31; 27 3.8 29.3 as (26) LA Male 22.9 (nfa) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa nis Female 4.8 (nfa) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 3?0 Both 14.5 (nia) 1310) 18.9 3.5) 18. 0.4) 2.2 31: 21 4.0 (25) 2?.4 ?it; (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nfa) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nfa) 10.3 0.3) 10.1 0.2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(0.2) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nfa) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 900 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (We) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 as (24) Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nfa) 4.8 0.0.9) 2.8 3?0 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51 Differences between ranks do not necessarily imply a statistically signi?cant difference. 1 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -3- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State REnk} Both 10.71n1a1 11.51+ 0.91 12.41+ 0.51 12.91+ 0.51 14.21+ 1.4.31191 40.5 fit-r1 51 MN Male 15.3 1n1a1 1931+ 1.11 20.41+ 1.01 2091+ 0.51 22.9 81+ 1.91 2331+ 0.41 1.7 ?fa 1p<.011 Female 3.5 1n1a1 4.21+ 0.51 51+ 0.51 1 1+ 0.41051 591+ 1.21 4.2 ?fa 1p<.011 Both 12.91n1a1 14.1 (+1.21 14. 7 1+ 0 51 1551+ 0.51 1551+ 0.11 15. 210 .31 +1.1 3011:11051 35 2.31351 17.8 ?55 1401 M5 Male 22.9 1n1a1 24.51+1.71 25.1 1+ 0. 51 2551+ 1.71 25.91.91 25.31.51 0.7 Female 4.3 1n1a1 501+ 0.7151+ 0. 51 51 0. 01 541+ 09121-021 2.4 '11: 1p<.011 Both 14.71n1a1 14.1 1? 0 51 1541+ 131 1501+ 0. 71 1751+ 1.71 20. 0 1+ 2.3] 2.2 i5131p<.011 15 5.31151 35.4 15117} MD Male 25.3 1n1a1 2371+ 1.51 25. 5 1+ 1.91 2551+ 1.01 2591+ 2.31 32. 2 1+ 3.31 1.5 '15 1p<.051 Female 5 4 (ma) 5.41+ 0.1111+ 0. 71 31+ 0.21 7.41+ 1.11 51+ 1.21 3.2 14113401} Both 21.1 1n1a1 22.51+ 1.41 2351+ 101 24.71+1.11 2571+ 2. 01 2921+ 2.51 2.1 1 5.01 21 35.0 351111 MT Male 35.9 1n1a1 3731+ 0.41 39. 5 1+ 2. 51 39.710 11 41 0.1+ 1 .41 4551+ 4.41 1.3 ?fa 1p<.011 Female 5.71n1a1 5.41+1.5141+0.11 10. 0 1+ 1. 51 1251+ 2.51 1311+ 0.51 4.5 551131011 Both 12.71n1a1 122111.71.51 1351+ 1.51 14.51+1.31 +1.0 ?fa 40 2.1 1421 15.2 ?11} 1431 NE Male 22.21n1a1 20.71- 1 51 20 310.2391+ 19.1 0.5 ?fa nis Female 3.5 1n1a1 4.21+ 0.4111+ 0.9101+1.2.5 Both 23.31n1a1 22.51- 0 51 2221.4 21 23.1 1+ 1.51 0.2 ?it. 9 0.21511 +1.0 51.1511 NV Male 35.3 1n1a1 3571+ 1.71 35.1 1+ 1.51 35510.7 51; his Female 5.91n1a1 951+ 0.5151+ 0.11 1001+ 0 .41 10. 5 1+ 0.61 11.21+ 0. 51 +1.5 ?fa 1p<.011 Both 13.51n1a1 12.51 .01 13.31+051 15.21 1. 91 15.51+0.51 20.0 {+4.21 +2.7 501p<.051 17 +5.51 5) +453 %1 31 NH Male 22.5 1n1a1 21.1 1+1 4) 21.71+ 0. 51 24. 5 1+ 3.2.2 551131051 Female 5.3 1n1a15+01.5191+ 1.01213.9 501p<051 Both 7.5 1n1a1 71+01151+ 0.2101.41 1.3 31: 1p<.051 50 1.51471 19.2 351351 NJ Male 13.01n1a1 13.11+ 0.01 12 510.551 13. 71+ 1.1 1 1451+ 0.51 1451+ 0.11 0.9 55151051 Female 3.2 1n1a1 91+ 0.31 3.01+ 0.0191 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; n15 indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 20011 and last (2014 2015) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -4- Trends in Suicide Rates among Persons 2 19 Years of Age, by State and Sex, 1999 2916 Age-Adjusted Annual Rate per 199,999 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2991 2992 2994 2995 2997 2999 2919 2911 - 2913 2914 2916 Rank 5 {State Rank) 17 (State ank) 55111 22.91n1?a) 22.014 01) 51 02) 2301+ 1.2) 24.11+1.1) 2501+ 1.9) +1.1351p<.05) 4 4.0124) 15.3 35139) NM Male 36.6 1nfa1-1 .2) 35.61 9. 6) 37.1 1+ 1 .3) 49.71+ 3.6) 9.4 35 nis Female 5.5 1n7a) 411.1) 20111.71+ 0. 9) 1201+ 0.3) 3.3 35 11:14.05) 55111 7.21n7a) 110.1+1. 1) 310 .1) 213515401) 49 2.1 141) 25.5 35127) NY Male 12.51n1a) 12.21413) 1291+ 0.7) 1391+ 1. 0) 15.41+ 1.4) 14.5019) 1.4 351p<.05) Female 2.71nfa) 1-9 .1) 391+ 93) 1+ 9. 5) 4.2 1+9. 7) 461+ 9.5) +4.2 35 1p<.91) 55111 13.51nia) 13.4)51 .1) 1371+ 0.1) 14.21+05) 14.5.1+04) 15..31+55) +0.5 3515401) 34 1.7144) 12.7 35 147) 1111; Male 22.7 1n7a) 2271+ 0.0) 22.2 1- 0.5) 2331+ 1.1) 2331n75 Female 5 6 11115) 51 9 2) 621+ 9.2.9 3511:1995) Both 13.31n1a) 1461+ 1.3) 1691+ 1.4) 16. 6 1+ 9 6) 16.41+ 1.9) 29. 9 1+ 2. 5) 2.9 351p<.91) 14 7.61 5) 57.6 351 1) ND Male 21.4 1n1a) 2451+ 3.2) 230129.11:14.01) Female 5.5 1n7a) 51? 1.0) 371- 0.1.0) 551+ 1 .5) 3.9 35 n75 55111 11.5 11115) 1231+ 0.5) 1311+0.5) 1341+ 0.2) 14.51+ 14.) 15.51+1.0) +2.0 35 15401) 32 +4.2121) +350 35119) OH Male 20.4 1n1a) 2091+ 0.5) 2221+ 1.3) 221.3) 1.5 35 1p<.01) Female 4.0 1n1a) 471+ 0.7) 491+ 0.1) 31+ 0. 5) 0.11:14.01) Both 17.91nta) 16. 51116.41+1.1) 2971+ 2. 3) 23.51+ 2.6) 2.3 35111995) 7 6.4119) 37.6 i75112) OK Male 26.5 1nfa) 27. 311 .2) 2761+ 9. 5) 3931+ 2.5) 3341+ 3.1) 3731+ 3.6) 2.9 3511:1995) Female 6.6 1nfa) 19 .2) 7. 5 1+ 1.1) 91- 9.5) 6.51+ 1 6.) 19.31+ 1.6) 2.9 1p<.95) Both 16.41nta) 17.71+1.3) 17. 71-9 .9) 16. 61+ 9.9) 19. 6 1+ 1 .2) 21.1 +1.6 351p<.91) 13 4.6116) 26.2 i55126) OR Male 27.41n7a) 2951+ 2.1) 25.51 0. 9) 29.51+1.0) 31.41+1.5) 33 01+ 1.5) 1.1 35 11:14.01) Female 6.5 11115) 1 1+ 9.6) 7.71+ 9.6) 41+ 9.7) 661+ 9.4) 9. 61+ 9. 9) 2.7 3519491) Both 12.1 1n1a) 1251+ 9.4) 1261+ 9.3) 13. 9 1+ 1.1) 15. 9 1+1. 1) 16.31+1.2) 2.9 351p<.91) 39 4.1 122) 34.3 35121) PA Male 21. 01n nfa a) 2131+ 0.3) 21 .91+05) 231 1+ 1.2) 2471+ 17.) 25.11+1.3) 15351121401) Female 4.2 1n7a) 451+ 0.3) 451+ 0.0) 41+ 0.9) 5. 0 1+ 0. 5) 11+ 1.1) 3.5 35 11:14.01) Rates are age-adjusted to the U5. year 2999 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 159 states and the District of Columbia) is for the reporting period 2914 2916. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2991) and last (2914 2916) reporting periods. Ranks are from largest increase 11) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 29 suicides. Percentage of injury deaths for which intent was not determined exceeded 29% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -5- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) a, Both 9.4 (nfa11.9 (.3.2 (30) 34.1 ?Va (23) RI Male 15.4 (Ma) .2) 14.8[- 0. 3) 1912?: 2.0) +2.2 ?fa nis Female 4.0 (nfa) 3-0( .7) 0. 4) 1. 3) 00) 6.1 1.0) 3.7 ?fa Both 12.8 (nfa) 13.0 0.2) 1316.0 1.1) 17.7 1.7) 2.3 ?fa 23 4.9 (17) 38.3 ?15(10) SC Male 21.3 (nfa) 22. 5 .2) 22.3 (.1) 24. 6 2. 2) 26.1 1.5) 28. 0( 1 .9) +1.8 ?fa Female 5 4 (Na) 7(-0 .7) 1.3) 0. 2) 7.0 0.8) 8. 4 1.4) 3.4 ?fa Both 15.7 (nfa) 1) 17.1 19.7(+ 0.4) 22.6 +2.5 ?fa 10 7) +445 ?Va( 6) SD Male 27.6 (We) 26.3 (.3) 27. 9 1 .6) 30.1 2. 2) 32.0 1.9) 33. 6 1 .6) 1.6 ?fa Female 4.2 (nfa) 5.8 1.5.8 ?fa Both 14.6 (nfa) 15.2 0.6) 1617.2 0.0) 18. 2 1 .0) +1.4 ?fa 22 3.5 (28) 24.2 ?Va (31) TN Male 25.1 (nfa) 25.4 0.3) 2628.6 0 6) 29. 8 1 .2) 1.2 ?fa Female 5.4 (his) 6.3 0.0.6) 7. 6 0. 7) 1.9 ?fa Both 12.2 (nfa) 12.7 0.6) 12.3 (- 0.14.5 0. 9) 1.1 ?fa 41 2.3 (37) 18.9 ?Va (36) TX Male 20.4 (Ma) 20.9 0.5) 20.4 (- 0.23.1 0. 9) 0.9 ?fa Female 4.8 (nfa) 5.4 0.6) 5.0 (a 0.1.6 ?fa Both 17.2 (nfa) 19.0 1.8) 18.2 0.7) 20. 2 2.0) 24. 0 611+ 3. 8) 25.2 1.2) 2.7 ?fa 5 8.0 3) 46.5 ?if? 4) UT Male 28.2 (nfa) 31.1 2.9) 29401.7) 32 5(+2.1 ?fa Female 6.8 (nfa) 7.4 0.6) 7.5 0.1) 10. 6 2.1) 12.6 2.0) 4.4 ?fa Both 13.2 (nfa) 16.2 14.9 (- 1.3) 16.6 1.7) 18.7 2.1) 19.7 1.0) 2.4 ?fa 18 6.4( 9) 48.6 ?Va( 2) VT Male 23.6 (We) 28. 3 .6) 24.3 (- 4.0) 27.3 3.0) 31.0 3. 7) 32.5 1.5) 1.9 ?fa Female 4.3 (nfa) 2H4 0. 9) 6.4 1.3) 6 6 0.2) 7. 31+ 0. 7) 7.6 0.3) 3.8 ?fa Both 12.8 (nfa) 12.9 0.3) 13.6 0.7) 14. 6 0. 9) 15.0 0.5) +1.2 ?fa 37 2.2 (39) 17.4 ?Va (41) 17.4 Male 21.6 (nfa) 21.3 (- 21.0 (a 0.4) 22.5 1.5) 23. 6( 1 .2) 23.9 0.2) 0.9 ?fa (pa-<05) Female 5.3 (We) 5.2 0.1) 5.9 0.7) 5.6 (- 0.3) 6. 4 0 8) 6.9 0.5) 1.8 ?fa Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. -5- Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (1115) 15411781+ 1.0) 1.1 55113405) 24 2.8133) 18.8 c111137) WA Male 24.? (nla) 2521+ 5) 2527.1 1+ 1.1) 13.5 '11 Female 5.0 (ma) 8. 4 1+ 0 8) 01+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% his 11 +58113) +37.1 i?15114) WV Male 27.2 (nfa) 3011+ 2. 9) 5.81 1.+11% nis Female 5.3 (Na) 551+ 0.1) 31 0.5) 7. 8 1+ 2. 3) 581+ 2. 2) 3.7 ?it: nis Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps.01) Female 5.1 (nfa0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (his) 521+ 0.5) 41+ {152) 1+ 1.4) 1251+ 1. 9) 3.2 1p<.01) Rates are age-adjusted to the U.S. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ll Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst and last periods. Rani-ts are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1. Percentage Changes in Annual Suicide Rates (per 100,000, Age-Adjusted) 2014-2016 Compared Against 1999-2001 I: Decrease 1.0% I:Ilncrease 18.3% - Increase 18.8% - 29.3% - Increase 31.9% - 37.4% - Increase 316% - 57.8% Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted 0R11 {n=20,446} Problemf Mental Square (95% Cl] [95% Cl) (n=9,407} Health Problem (n=11,039] Sex Male 15,702l768} 9,233l83.6} Female 1,806l16.4] p<.01 DANA-0.5) Age" 10?24 1,593l14.4} p<.01 2544 45-64 3,898853] p<.01 65+ p<.01 White, non?Hispanic [0-ch Black, non-Hispanic Mil-4.4] 81717.4) p<.01 American Indian/Alaska Native, non- Hispanic 378(18} 112(12] 2669.4) Asian, non-Hispanic 576(28} 235(25} 3418.1) p<.05 Hispanic ?63019] 6336.7) Other 2103.2) 4s{0.4) Extended demographics Ever served in militarWr 2,075i20.l] Homeless 240(12} 104(1.1] 13641.3) incident Type Single suicide 20,063i982} 10343914) Homicide followed by suicide 319(1.6) 64(0.7} 255(23) p<.01 Multiple suicides 6483.3) 25(03} Method Firearm Ha 23671269] p<.01 Poisoning Substance class causing death?? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marij ua na Tested Positive Antidepressants Tested Positive 1.021(340) 944(314) 300(266) 6241208} 219013) 13,317l65.1} 3,554i41s) 499(6.3} 73619.7) 666(35.8} 603(322) 6441346} 468i25.1} 1951105) 230(33) 6,658i70.8) 4,258i453) 1,238i29.1) 3,866i41.1) 216(56) 376(102} 210(227) 355(311) 336(29.4) 156{13.7) 24(2.1) 315(14) 5.541(50.2) 3.398(353) 325(212) 4.1121372) 283(63) 3.919(355] 360(92) 3.442(312) 261(221) 2.322(210) 479{20.6) p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.Dl p<.01 05004-05) O.5i0.4-0.6] 0.210.103) 0.910.310) 1.111.012) 03(07-03) 1.211.015) 10(09?11) 09(0340) news-.99) 0.5035406) 0.940.910) 09400-10) Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina. Ohio, Oklahoma. Oregon. Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. had been identi?ed as having a current diagnosis of mental health problem in coronere?medical examiner or law enforcement reports. l2Odds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex, race and ethnicity. Known erlP was used as the reference group. were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged 18 years of age and older with reported military service status. Denominator is who died by poisoning, including overdose. ll? Denominator is with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted DR11 Problem* Mental Square [95% Cl] (95% Cl] Health Problem Suicide with known circumstances 18364913) 9,407i100) 9,357l84.8] p<.01 Mental Health Any Current Diagnosed Mental Health Problem? Anxiety disorder Bipolar disorder Schizophrenia 5096.4] PTSD 424{4.5l 226{2.4j Unknown 760{8.1] Current depressed mood p<.01 Substance Problems Any Current substance problem 5319(283) p<.01 Alcohol problem 3,268l17.4) 1,406l15.0] p<.01 Other substance problem 3,084l16.4) p101 Treatment Current mental abuse treatment 5,141l27.4) 64(0.7) p<.01 0.01} Ever treated for mental health/substance 0.02(0.02- problem 394(4.2} p<.01 0.02} Relationship Problems?oss Any relationship problem?oss 7,948t42.4) 3326896) 4,222l45.1) p<.01 Intimate partner problem 5,098l27.2) 2,828l30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131{1.4) 283(30} p<.01 Victim of interpersonal violence within past month 84(0.4) 53(0.6) 31(03) p<.05 Familyr relationship problem 8738.3) 798(35} Other relationship problem (non-intimate) 40342.1) 202l2.1) 201(2.1) Argument or conflict (not specified) 2,914l15.5) 1,636l17.5) p<.01 Death of a loved one (any) 826l8.8) 671(12} p<.01 Non?suicide death 534(57): p<.01 Suicide of family or friend 217(23) 162(1.7) p<.01 Other Life Stressors Any life stressor 9343619) 4,6?5l?93) Recent criminal legal problem 1,588l8.5) 586l6.2) p<.01 Other legal problem 748(4?) 373{4.0) Physical health problem 4.179(223) 2,167l23.2) p<.01 Jobg?Financial problem? 2941(16.2) 1530(16.8) 1411(15.6) p<.05 Eviction or loss of home 7228.8) 405(43} p<.01 School problem?? 162(193) 92(21.9) Recent release from an institution? 941(10.2) p<.01 0.5(0.4u0.5} Jailfprisonldetention facility 203(14.4) 121(25.7) Hospital 51786.6) 31183.0) 206043.?) p<.01 1.3i1.0-1.7) hospital/institution 46983.2) 4i39(46.7) 30(6.4) p?.01 Other (includes alc/SA treatment facilities) 223(15.8) 109(11.6) 114(24.2] p<.01 Recent or Impending Crisis Crisis within past or upcoming two 5,525l29.4) 2,444l26.0) 3,081l32.9) p<.01 Intimate partner problem crisis 1968856) 854(343) 1114(36.2) Physical health problem crisis 739(13.4) 315(12.9) 424(13.8] 1.003.842) Criminal legal problem crisis 621(11.2) 2031283) 418(13.6) p<.01 1.711.511} Family relationship problem crisis 4300.8) 2180.1) p105 Job problem crisis 354(6.4) 1910.8) 163(5.3) p<.01 Suicide Event/History Left a note assatsas) 3,182l33.8) 3,286t35.1) Disclosed suicide intent 4,405l23.5) 2,306l245) 2,099l22.4) p<.01 History of ideation 5,990l313) 3,838l40.8) 2,152l23.0) p<.01 History of attempts 3,732l19.9) 2,770l29.4) 962(103} [3101 *Alaska, Arizona. Colorado, Connecticut. (Lienrgia. Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey. New Mexico, New York, North Carolina. Ohio, Oklahoma. Oregon, Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis ofmental health problem in coronen?medical examiner or law enforcement reports. iiOdds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. includes decedents with one or more diagnosed current mental health problems. which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. Denominator is decadents aged 18 years of age and older. Denominator is decedents aged 10-18 years. '"15 Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. int? Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. April 10, 2018 Table 1. Trends in Suicide Rates among Persons 1: 19 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 109,999 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2091 2002 2904 2095 2007 2903 2010 2011 2913 2914 2015 Rank (State Rank) 11 (State Ragnk) H. Both 12.3 (We) 1271+ 0.4) 12.9 0.2) 13.31+ 0.9) 14.5 1+ 0.3) 1541+ 0.9) 1.5 c111154.01) nia 3.1 (Na) 25.4 a. (nia) U.S. Male 29.9 (nia) 2121+ 0.4) 2131+ 0.0) 22.5 23.51+1.0) 24.51+1.0) Female 4.7 5.0 1+ 0.3) 5.3 1+ 0.2) 5.7 1+ 0.4) 5.2 1+ 0.5) 5.9 1+ 0.7) 2.5 %1p<.01) Both 14.3 (nia) 13.41- 0.9) 14.1 1+ 0.5) 15.51+1.5) 15.41+ 0.7) 17. 51+ .1) +1.5 %1p<.05) 25 3.1131) 21.9 %133) AL Male 25.1 (nia) 23.4 (- 1.7) 24.41+ 1.0) 25.41+ 2.0) 27.5 1+ 1.1) 29.1 1+ 1. 5) 1. 3 ?/61 (cs. 05) Female 5.1 (nia) 4.31- 0.3) 501+ 0.2) 1 5.41+ 0.3) .01+ 0. 7) 2. 5 5'41 01) Both 21.0 (nia) 2431+ 3.3) 24.21- 0.5) 2501+ 1.7) 25.41- 0.5) 2331+ 3.4) +1 7 (psi 05) 2 7.31 4) 37.4 113) AK Male 33.2 (nia) 3311+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 0.1) 42. 9 1+ 2. 3) +1.4 1p<. 01) Female 3.51nia) 11.41+ 2.9) 9.31-1.5) 11.1 1+ 1.2) 991-12nis Both 17.31nia) 1351+ 0.7) 19.1 1+ 0.5) 19.11- 0.0) 20. 4 1+ 1.3) 20. 9 1+ 0. 5) +1.0 %11.p< 01) 15 3.1132) 17.3 142) .42 Male 29.3 (nia) 3021+ 1.0) 30. 5 1+ 0.4) 30. 2 1? 0.5) 32. 0 1+ 1.1pc. 05) Female 7.1 (nia) 7.51+ 0.4) 21+ 0.7) 51+ 0.5) 21+ 0.5) 91+ 0. 5) 2. 2 1p<. 01) Both 15.5 We) 1531+ 0.3) 15.21+0.5) 1751+ 1.4) 1921+ 1.5) 21 g..21+20) 01) 12 +5.7114) +35.3%115) AR Male 25 71nia) 2571+ 0.0) 27. 2 1+ 0.5) 23. 2 1+ 1.0) 31. 7 1+ 3.5) 33. 5 1+ 1. 9) 1.5 5'41 1pc. 05) Female 5.5 (nia) 5.91+ 0.3) 5. 2 1+ 0 4) 91+ 1.7) 7.51- 0.4) 51+ 2.1) 3. 5 31:1 01) Both 10.5 (nia) 11.3 11 .01- 03) 12.01+1.0) 11.31- 0.1) 121 05) 45 1.5145) +14.3%145) CA Male 17.9 (nia) 1341+ 0.5) 17.7-( l0.7) 19.1 1+ 1.4) 13.9 (-0.2) 19.21+0.3) +05% nis Female 4.1 (nia) 501+ 09) 4. 0. 1) 41+ 0.5) 5. 3 1- 0.1) .51+ 0. 3) 7 5'41 1pc. 05) Both 17.31nia) 19.21+1.9) 19. 01-0 2) 20. 0 1+ 1.0) 21.51+1.5) 2321+ 15) 1.3 (psi 01) 3 5.9112) 34.1 122) CO Male 23.5 (nia) 3091+ 2.3) 30.510 .4) 31.51+1.0) 33.41+1.9) 35. 3 1+ 2. 9) +1.4 1p<. 01) Female 7.0 (nia) 3.21+1.3) 321+ 0.0) 11+ 0.9) 10.1 1+ 1.?/61 (cs. 01) Both 9.5 (nia) 3.91? 0.7) 9.1 1+ 0.2) 10. 2 1+ 1.1) 11.01+ 0.3) 1151+ 0.5) +1.5 (cs. 05) 45 1.9143) +192 11211134) CT Male 15.4 (nia) 14.5 1? 1.3) 1501+ 0.4) 15. 5 1+ 1.5) 17. 5 1+ 1.0) 17.31- 0.3) 0.9 his Female 3.5 (nia) 331+ 0.2) 3.71- 0.2) 41+ 0.7) 91+ 0.5) 5.21+ 1.3) 3.5 %1p<.05) Rates are age-adjusted to the US. year 2000 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest percentage increase 11} to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 20 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank} 1? (State REnk} Both 13.6 1n1a1 12.21.41 1191? 0.31 13.61+ 1.71 14.21+ 0.61 14.41+ 0.21 0.9 nis 42 0.81501 5.9 ?it: 1501 DE Male 23. 0 1n1a1 20.31.71 1991- 0.41 23.1 1+ 3.21 22. 71- 0. 41 2351+ 0.81 0.6 ?fa Female 31ni'a1 010 .21 4.61? 0.41 91+ 0.31 6. 41 1.51 21- 0.21 1.6 nl's Both 91nia1 6.41+ 0.51 6.41? 0.01 31+ 0.81 6?01.71 6.91+ 0.31 0.9 ?fa 51 1.01481 16.1 ?it 1451 DC Male 10.71n1a1 11.11+ 0.41 10.31; 0.81 12.71+ 2.41 10. 012 .61 11. 71+ 1.71 0.3 ?it: nis Female 1.71nia1?FT 2.31+ 0.611?r 331+ 1.01 61- 0.71 3.61+ 1018-0181 3.5 ?fa Both 14.81n1a1 1521+ 0.41 14.91- 0.31 16. 3 1+ 1.41 16. 31?0 .1 16.41+ 0.11 0.8 301p<.051 29 1.61451 10.6 i131481 FL Male 24.3 1n1a1 24.41+ 0.11 23.61- 0.81 26. 2 1+ 2.61 25. 61 0. 61 25.6 1?0. 11 0.5 ?it. Ms Female 6 3 (ma) 681+ 0.51 6.81+ 0.0111+ 0.31 7.61+ 0.51 81+ 0.31 1.4 301p<011 Both 12.91n1a1 13.21+ 0.31 12.31? 0.91 13. 2 1+ 0.91 13.71+ 0.51 1501+ 1.31 0.9 ?it; nits 39 2.1 1401 +162 131441 GA Male 22.1 1n1a1 23.1 1+ 1.01 21311.81 .91+ 0.61 2261+ 0.71 24.-41+ 1.71 0.5 ?fa nl's Female 5.0 1n1a1810.21 4.6 1- 0.2151+ 0.91 5.8 1+ 0.31 6.6 1+ 0.81 2.1 ?fa 1p<.051 Both 12.91nla1 11. 1 11.81 10.31- 0.71 14. 5 1+ 4.11 14.41? 0.11 1521+ 0.81 2.0 ?fa 35 2.41351 18.3 I1111381 HI Male 20.4 1n1a1 17.21.11 15.3 1- 1.91 21. 91+ 6.71 2251+ 0.51 24.31+ 1.81 2.1 ?fa Female 5.4 1n1a1010.41 551+ 0.5111+ 1.51 21- 0 91 5.91- 0.31 1.2 ?fa nl's Both 17.31n1a1 1921+ 2.01 18.31? 0.91 21. 61+ 3.31 2191+ 0.31 24.71+ 2.81 2.3 3611:1101} 6 7.51 61 43.2 1 71 ID Male 28.4 1n1a1 33.1 1+ 4.71 31.1 1? 2.01 3 91+ 3.81 71- 0 21 3801+ 3.31 1.6 301p<051 Female 7.2 1n1a1 6.1 1- 1.11 6.1 1+ 0.0101+ 2.91 9.51+ 0.51 11.81+ 2.31 4.4 1p<.051 Both 9.9 1n1a1 9.81? 0.11 9.71- 0.11 10. 6 1+ 0.81 11.21+ 0.61 1221+ 1.01 1.5 3131p<.051 44 2.31381 22.8 ?it 1321 IL Male 17.1 1n1a1 16.71? 0.41 16.2 1- 0.41 17. 61+ 1.41 18.51+ 0.91 1981+ 1.31 1.1 36113105) Female 3.7 1n1a1 3.814 0.01 381+ 0.2121+ 0.41 4.51+ 0.41 521+ 0.61 2.4 301p<011 Both 13.01n1a1 13.71+ 0.71 14.41+ 0. 71 1491+ 0.51 16.41+1.41 17.11+ 0.71 +1.9 31: 1p<.011 26 4.11231 31.9 13125} IN Male 22.4 1n1a1 23.21+ 0.81 24. 41+ 1.21 24. 71+ 0.41 2671+ 2.01 28.31+ 1.61 1.5 36113101} Female 4.6 1n1a1 501+ 0.41 531+ 0.2191+ 0.61 681+ 0.91 6.61? 0.21 2.7 361131011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia1 is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 (-1.4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nia) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 i1110.14.01) 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nia) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nia) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia.4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 ?it; 27 3.8 29.3 14(26) LA Male 22.9 (nia) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa Female 4.8 (nia) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 31;? Both 14.5 (nla) 1310) 18.9 3.5) 18. 0.4) 2.2 51: 21 4.0 (25) 2?.4 ?11} (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nia) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nia) 10.3 0.3) 10.1 0. 2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(02) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nia) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 90(4 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (nia) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nia) 4.8 0.0.9) 2.8 31;? Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State ank) Both 10.71n1a) 11.51+ 0.9) 12.41+ 0.5) 12.91+ 0.5) 14.21+1.3) 15. 01 0. 9) 2.3 701p<.01) 35 4.3119) 40.5 fit-r1 5) MN Male 15.3 1nia) 1931+ 1.1) 20.41+ 1.0) 2091+ 0.5) 22.9 81+ 1.9) 2331+ 0.4) 1.7 ?fa 1p<.01) Female 3.5 1n1a) 4.21+ 0.5) 51+ 0.5) 1 1+ 0.4) 05) 591+ 1.2) 4.2 55113101) Both 12.91nia) 14.1 1+ 1.2) 14. 7 1+ 0. 5) 1551+ 0.5) 1551+ 0.1) 15. 210 .3) +1.1 3511:1105) 35 2.3135) 17.8 i55140) M5 Male 22.9 1n1a) 24.5 1+ 1.7) 25.1 1+ 0 5) 2551+ 1.7) 25.910 .9) 25. 31.5) 0.7 nlis Female 4.3 1nia) 501+ 0.541+ 0.9) 210.2) 2.4 1p<.01) Both 14.71n1a) 14.1 1? 0 5) 15.41+1.3) 1501+ 0. 7) 17.51+ 1.7) 20. 0 1+ 2.3) 2.2 15 5.3115) 35.4 i15117) MD Male 25.3 1nia) 23.71? 1.5) 25. 5 1+ 1 .9) 2551+ 1.0) 2591+ 2.3) 32. 21+ 3.3) 1.5 501p<.05) Female 5 41n1a) 5.41+ 0.1) 11+ 0. 7) 31+ 0.2) 7.41+ 1.1) 51+ 1.2) 3.2 501p<.01) Both 21.1 1nia) 2251+ 1.4) 23. 5 1+ 1 .0) 24.71+1.1) 2571+ 2.0) 29.21+ 2.5) 2.1 1 5.01 2) 35.0 35111) MT Male 35.9 1n7a) 3731+ 0.0.1) 41 0.1+ 1 .4) 4551+ 4.4) 1.3 55113101) Female 5.71nia) 5.41+1.5) 41 0.1) 10. 0 1+ 1. 5) 12.51+ 2.5) 1311+ 0.5) 4.5 3511:1101) Both 12.71ni'a) 1221? 0 5) 12. 51+ 0 .4) 11. 71?0 5) 1351+ 1.5) 14.51+1.3) +1.0 ?fa 40 2.1 142) 15.2 i111143) NE Male 22.21nia) 2071-1523.91+19.) 0.5 ?fa nis Female 3.5 1n7a) 4.21+ 0.4)11+ 0 9) 01 1.2.5 ?70 his Both 23.31nia) 22.51- 0 5) 22.1 1 0. 5) 2251+ 0. 5) 21.4 2) 23.1 0.2 ?it. n15 9 0.2151) -1.0 55151) NV Male 35.3 1n1a) 35.714 1.7) 35.1 1 1 .5) 35510.7 51; 1115 Female 5.91nia) 951+ 0.5) 51+ 0.1) 100111.21+ 0. 5) +1.5 ?fa 1p<.01) Both 13.51n1a) 12.51?1.) 13.31+0.5) 15.21 1. 9) 15.51+0.5) 20.0 +2.7 3131p<.05) 17 +5.51 5) +453 %1 3) NH Male 22.51n1a) 21.1 11.4) 21.71+ 0. 5) 24. 5 1+ 3.2.2 5511:1105) Female 5.3 1n1a) 510.5) 91+ 1.3.9 501p<.05) Both 7.5 1n1a)7?01 1) 51 0.2 01+ 0.551.5147) 19.2 35135) NJ Male 13.01n1a) 13.11+ 0.0) 12.51 0.55) 13. 71+ 1.1 14.51+ 0.5) 1451+ 0.1) 0.9 %1p<.05) Female 3.2 1nia) 91? 0.3) 301+ 0.0) 91 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2006 2010 2011 - 2013 2014 2016 Rank 5 {State Rank) 17 (State REnk} 55m 22.01n1a1 2201- 011510.21 2301+ 1.21 24.1 1+ 1.11 25.01+1.91 +1.1351p<.051 4 4.01241 15.3 351391 NM Male 36.6 1nfa1 3737.1 1+ 1 .31 4071+ 3.61 0.4 35 nis Female 5.5 1n1a1411.1121011.71+ 0. 91 1201+ 0.31 3.3 351114.051 Both 7.2 1n1a1 110 .1+1. 11 310 .11 21351114011 49 2.1 1411 25.5 351271 NY Male 12.51n1a1 12.210.31 1291+ 0.71 1391+ 1. 01 1541+ 1.41 14.510 .51 1.4 351p<.051 Female 2.71nfa161-0.11 301+ 0311+ 0. 51 4.2 1+0. 71 461+ 0.51 +4.2 35 1p<.011 55m 13.61nl'a1 13510 .11 1371+ 0.11 14.21+0.51 14.5.1+041 15..31+051 +0.5 35154011 34 1.71441 12.7 35 1471 no Male 22.7 1n1a1 2271+ 0.01 22.2 1- 0.51 2331+ 1.11 2331n75 Female 5 61nfa1 51- 0 21 621+ 0.2.0 35113905} Both 13.31nl'a1 14.61+1.31 1601+ 1.41 16. 6 1+ 0. 61 1641+ 1.91 20. 9 1+ 2. 51 2.9 351p<.011 14 7.61 51 57.6 351 11 no Male 21.4 1n1a1 2451+ 3.21 23012.5 351114.011 Female 5.5 1n1a1 51- 1.01 371- 0.5171+ 2 01 5. 7 1+ 1.01 551+ 1 5.1 3.9 35 n75 Both 11.5 1n1a1 1231+ 0.51 131 1+ 0.51 1341+ 0.21 14.51+ 1.41 15.51+1.01 +2.0 35 1p<.011 32 +4.21211 +350 351191 OH Male 20.4 1n1a1 2091+ 0.51 2221+ 1.31 221p<.011 Female 4.0 1n1a1 471+ 0.71 491+ 0.1131+ 0.51091 5. 7 1+ 0. 51 3.4 351114.011 Both 17.0 1n1'a1 16. 511641+ 1.11 20. 7 1+ 2 31 2351+ 2.61 2.3 35109051 7 6.41101 37.6 i151121 OK Male 26.5 1nfa1 27.31.21 2761+ 0. 51 3031+ 2.51 3341+ 3. 11 3731+ 3.61 2.0 351139051 Female 6.6 1nfa1401- .21 7. 5 1+ 1.11 01- 0.51 651+ 1.61 10.31+ 1.61 2.9 1p<.051 Both 16.41nta1 17.71+1.31 17.71.01 16. 61+ 0.91 19. 6 1+ 1 .21 21.1 {+1.31 +1.6 351p<.011 13 4.61161 26.2 i551261 OR Male 27.41nl'a1 2951+ 2.11 25.51 0. 91 29.51+1.01 31.41+1.51 33 01+ 1.51 +1.1 351154.011 Female 6.51nfa1 1 1+ 0.61 7.71+ 0.6141+ 0.71 6612.7 35109011 Both 12.1 We] 1251+ 0.41 12.61+ 0.31 13. 9 1+ 1.11 15. 0 1+1. 11 16.31+1.21 2.0 351p<.011 30 4.1 1221 34.3 351211 PA Male 21. cm n1a a1 2131+ 0.31 21 .91+051 231 1+ 1.21 2471+ 17.1 25.11+1.31 1.5 351114.011 Female 4.2 1nl'a1 451+ 0.31 451+ 0.0141+ 0.91 5. 0 1+ 0.5111+ 1. 11 3.5 35 1114.011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period1" Current Overail Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 17 (State REnk} Both 9.41n1a10103101 0.0 01 1281+ 3. 81 11.91.91 12. 6 1+ 0. 71 2.6 ?fa 1p<.051 43 3.2130 W1 34.1 ?fa 123 W1 RI Male 15.41n7a1 1521-0 .21 14.81- 0. 31 21.21+6.41 1912?: 2.01 19..61+041 +2.2 ?fa Female 4.0 1n1a1310.7181+ 0. 41 11+ 1.31001 6.1 1+ 1 .01 3.7 ?fa 1p<.051 Both 12.81n1a1 13.01+ 0.1601+ 1.11 17.71+1.71 2.3 ?fa 1p<.011 23 4.91171 38.3 ?161101 SC Male 21.31n1a1 22.5 1+ 1.21 22.31.11 24. 6 1+ 2. 21 26.11+1.51 28.01+1.91 +1.8 ?fa 1p<.011 Female 5 4 [n7a171-0.7101+ 1.3121+ 0. 21 701+ 0.81 8. 4 1+ 1.41 3.4 ?fa 1p<.051 Both 15.71n1a1 15.81+0. 11 17.1 1+ 1.31 19.31+2.21 1971+ 0.41 22.6 +2.91 +2.5 ?fa 1p<.011 10 +7.01 71 +44.5%1 61 SD Male 27.6 [n1a1 26. 311 .31 27. 9 1+ 1.61 30.1 1+ 2. 21 3201+ 1.91 33. 6 1+ 1.61 1.6 ?fa 1p<.011 Female 4.2 1n1a1 5.81+1.6141+ 0 6131+ 2. 01 31-1 01 11. 31 4. 01 5.8 ?fa 1p<.011 Both 14.61n1a1 1521+ 0.61 16111721+ 0.01 18. 2 1+ 1 .01 +1.4 ?fa 1p<.011 22 3.51281 24.2 ?161311 TN Male 25.1 1n1a1 2541+ 0.31 2681+ 1.31 8 01+ 1. 21 2861+ 0 61 29. 8 1+ 1.21 1.2 ?fa 1p<.011 Female 5.4 [n1a1 631+ 0.91 6 71+ 0 4151+ 0. 81 6.9 (a 0.61 7. 6 1+ 0. 71 1.9 ?fa 1p<.051 Both 12.21nfa1 12.71+ 0.61 12.31- 0.41 13211451+ 0. 91 +1.1 ?fa1p<.011 41 2.31371 18.9 ?161361 TX Male 20.4 [n7a1 2091+ 0.51 20.4 1- 0.61.0123.1 1+ 0. 91 0.9 ?fa 1p<.051 Female 4.8 [his] 5.4 1+ 0.61 5.0 1+ 0.4121+ 0. 21 0.41 6. 4 1+ 0. 81 1.6 ?fa 1p<.051 Both 17.21n1a1 19.0 1+ 1.81 18.21? 0.2521+ 1.21 2.7 ?fa 1p<.011 5 8.01 3 46.5 ?fa 1 4 1W1 UT Male 28.21n1a1 31.1 1+ 2.91 29.4 1- 1.71 32 512.1 ?fa 1p<.051 Female 6.8 [n7a1 7.41+ 0.61 7.51+ 0.2.11 1261+ 2.01 4.4 ?fa 1p<.011 Both 13.21n1a1 16.2 (+3.01 14.91-131 16.61+1.71 18.71+ 2.11 19.71+ 1.01 2.4 ?fa 1p<.011 18 6.41 91 48.6 ?Va 1 21 VT Male 23.6 [n1a1 28. 31+ 61 24.31- 4.01 2731+ 3.01 3101+ 3. 71 3251+ 1.51 1.9 ?fa 1p<.051 Female 4.3 1n1a1 21+4 0. 91 641+ 1.31 6 61+ 0.21 7. 31+ 0. 71 761+ 0.31 3.8 ?fa 1p<.011 Both 12.81n1a1 12.71-011 1291+ 0.31 1361+ 0.71 1461+ 0. 91 1501+ 0.51 +1.2 ?fa 1p<.011 37 2.21391 17.4 ?161411 17.4 Male 21.6 1n1a1 21.31- 1 2101? 0.41 2251+ 1.51 23.61 1 2.1 2391+ 0.21 0.9 ?fa 1p<.051 Female 5.3 [n1a1 5.21? 0.11 591+ 0.71 5.61? 0.31 6. 4 1+ 0 81 691+ 0.51 1.8 ?fa 1p<.051 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (nia) 15411781+ 1.0) 1.1 5511:1405) 24 2.8133) 18.8 c15137) WA Male 24.? [nfa) 2521+ 0 5) 2527.1 1+ 1.1) 13.5 '14 Female 5.0 (We) 8. 4 1+ 0 8) 81+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% 11 +58113) +37.1 i?15114) WV Male 27.2 (nia) 3011+ 2. 9) 5.81 1.+11% Female 5.3 (Na) 551+ 0.581+ 2. 2) 3.7 ?it: Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps01) Female 5.1 (nia0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We.2) 47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (ma) 521+ 0.5) 41+ 5 52) 1+ 1.4) 1251+ 1. 9) 3.2 '14. 1p<.01) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Changes in Annual Suicide Rates (per 100,000, Age?Adjusted) 2014-2016 Compared Against 1999-2001 :Decrease 1.0% I:1lnr:rease 59% - 18.3% - Increase - 29.3% - Increase 31.9% - 37.4% - increase 316% - 57.8% Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted 0R11 {n=20,446} Problemf Mental Square (95% Cl] [95% Cl) (n=9,407} Health Problem (n=11,039] Sex Male 15,702l768} 9,233l83.6} Female 1,806l16.4] p<.01 DANA-0.5) Age" 10?24 1,593l14.4} p<.01 2544 45-64 3,898853] p<.01 65+ p<.01 White, non?Hispanic [0-ch Black, non-Hispanic Mil-4.4] 81717.4) p<.01 American Indian/Alaska Native, non- Hispanic 378(18} 112(12] 2669.4) Asian, non-Hispanic 576(28} 235(25} 3418.1) p<.05 Hispanic ?63019] 6336.7) Other 2103.2) 4s{0.4) Extended demographics Ever served in militarWr 2,075i20.l] Homeless 240(12} 104(1.1] 13641.3) incident Type Single suicide 20,063i982} 10343914) Homicide followed by suicide 319(1.6) 64(0.7} 255(23) p<.01 Multiple suicides 6483.3) 25(03} Method Firearm Ha 23671269] p<.01 Poisoning Substance class causing death?? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marij ua na Tested Positive Antidepressants Tested Positive 1.021(340) 944(314) 300(266) 6241208} 219013) 13,317l65.1} 3,554i41s) 499(6.3} 73619.7) 666(35.8} 603(322) 6441346} 468i25.1} 1951105) 230(33) 6,658i70.8) 4,258i453) 1,238i29.1) 3,866i41.1) 216(56) 376(102} 210(227) 355(311) 336(29.4) 156{13.7) 24(2.1) 315(14) 5.541(50.2) 3.398(353) 325(212) 4.1121372) 283(63) 3.919(355] 360(92) 3.442(312) 261(221) 2.322(210) 479{20.6) p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.Dl p<.01 05004-05) O.5i0.4-0.6] 0.210.103) 0.910.310) 1.111.012) 03(07-03) 1.211.015) 10(09?11) 09(0340) news-.99) 0.5035406) 0.940.910) 09400-10) Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carol ina, Ohio, Oklahoma. Oregon. Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. ?Ddds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. Dcecdents were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged 13 years of age and older with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted DR11 Problem* Mental Square [95% Cl] (95% Cl] Health Problem Suicide with known circumstances 18364913) 9,407i100) 9,357l84.8] p<.01 Mental Health Any Current Diagnosed Mental Health Problem? Anxiety disorder Bipolar disorder Schizophrenia 5096.4] PTSD 424{4.5l 226{2.4j Unknown 760{8.1] Current depressed mood p<.01 Substance Problems Any Current substance problem 5319(283) p<.01 Alcohol problem 3,268l17.4) 1,406l15.0] p<.01 Other substance problem 3,084l16.4) p101 Treatment Current mental abuse treatment 5,141l27.4) 64(0.7) p<.01 0.01} Ever treated for mental health/substance 0.02(0.02- problem 394(4.2} p<.01 0.02} Relationship Problems?oss Any relationship problem?oss 7,948t42.4) 3326896) 4,222l45.1) p<.01 Intimate partner problem 5,098l27.2) 2,828l30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131{1.4) 283(30} p<.01 Victim of interpersonal violence within past month 84(0.4) 53(0.6) 31(03) p<.05 Familyr relationship problem 8738.3) 798(35} Other relationship problem (non-intimate) 40342.1) 202l2.1) 201(2.1) Argument or conflict (not specified) 2,914l15.5) 1,636l17.5) p<.01 Death of a loved one (any) 826l8.8) 671(12} p<.01 Non?suicide death 534(57): p<.01 Suicide of family or friend 217(23) 162(1.7) p<.01 Other Life Stressors Any life stressor 9343619) 4,6?5l?93) Recent criminal legal problem 1,588l8.5) 586l6.2) p<.01 Other legal problem 748(4?) 373{4.0) Physical health problem 4.179(223) 2,167l23.2) p<.01 Jobg?Financial problem? 2941(16.2) 1530(16.8) 1411(15.6) p<.05 Eviction or loss of home 7228.8) 405(43} p<.01 School problem?? 162(193) 92(21.9) Recent release from an institution? 941(10.2) p<.01 0.5(0.4u0.5} Jailfprisonldetention facility 203(14.4) 121(25.7) Hospital 51786.6) 31183.0) 206043.?) p<.01 1.3i1.0-1.7) hospital/institution 46983.2) 4i39(46.7) 30(6.4) p?.01 Other (includes alc/SA treatment facilities) 223(15.8) 109(11.6) 114(24.2] p<.01 Recent or Impending Crisis Crisis within past or upcoming two 5,525l29.4) 2,444l26.0) 3,081l32.9) p<.01 Intimate partner problem crisis 1968856) 854(343) 1114(36.2) Physical health problem crisis 739(13.4) 315(12.9) 424(13.8] 1.003.842) Criminal legal problem crisis 621(11.2) 2031283) 418(13.6) p<.01 1.711.511} Family relationship problem crisis 4300.8) 2180.1) p105 Job problem crisis 354(6.4) 1910.8) 163(5.3) p<.01 Suicide Event/History Left a note assatsas) 3,182l33.8) 3,286t35.1) Disclosed suicide intent 4,405l23.5) 2,306l245) 2,099l22.4) p<.01 History of ideation 5,990l313) 3,838l40.8) 2,152l23.0) p<.01 History of attempts 3,732l19.9) 2,770l29.4) 962(103} [3101 *Alaska, Arizona. Colorado, Connecticut. (Lienrgia. Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey. New Mexico, New York, North Carolina. Ohio, Oklahoma. Oregon, Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis ofmental health problem in coronen?medical examiner or law enforcement reports. iiOdds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. includes decedents with one or more diagnosed current mental health problems. which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. Denominator is decadents aged 18 years of age and older. Denominator is decedents aged 10-18 years. '"15 Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. int? Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. April 10, 2018 Table 1. Trends in Suicide Rates among Persons 1: 19 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 109,999 Persons {Change from Prior Period) Current Overall Overall State Sex State Rate Change 2:229"; 1999 2091 2002 2904 2095 2007 2903 2010 2011 2913 2914 2015 Rank (State Rank) 11 (State Ragnk) H. Both 12.3 (We) 1271+ 0.4) 12.9 0.2) 13.31+ 0.9) 14.5 1+ 0.3) 1541+ 0.9) 1.5 c111154.01) nia 3.1 (Na) 25.4 a. (nia) U.S. Male 29.9 (nia) 2121+ 0.4) 2131+ 0.0) 22.5 23.51+1.0) 24.51+1.0) Female 4.7 5.0 1+ 0.3) 5.3 1+ 0.2) 5.7 1+ 0.4) 5.2 1+ 0.5) 5.9 1+ 0.7) 2.5 %1p<.01) Both 14.3 (nia) 13.41- 0.9) 14.1 1+ 0.5) 15.51+1.5) 15.41+ 0.7) 17. 51+ .1) +1.5 %1p<.05) 25 3.1131) 21.9 %133) AL Male 25.1 (nia) 23.4 (- 1.7) 24.41+ 1.0) 25.41+ 2.0) 27.5 1+ 1.1) 29.1 1+ 1. 5) 1. 3 ?/61 (cs. 05) Female 5.1 (nia) 4.31- 0.3) 501+ 0.2) 1 5.41+ 0.3) .01+ 0. 7) 2. 5 5'41 01) Both 21.0 (nia) 2431+ 3.3) 24.21- 0.5) 2501+ 1.7) 25.41- 0.5) 2331+ 3.4) +1 7 (psi 05) 2 7.31 4) 37.4 113) AK Male 33.2 (nia) 3311+ 4.9) 3391+ 0.3) 40.1 1+ 1.2) 40.1 1- 0.1) 42. 9 1+ 2. 3) +1.4 1p<. 01) Female 3.51nia) 11.41+ 2.9) 9.31-1.5) 11.1 1+ 1.2) 991-12nis Both 17.31nia) 1351+ 0.7) 19.1 1+ 0.5) 19.11- 0.0) 20. 4 1+ 1.3) 20. 9 1+ 0. 5) +1.0 %11.p< 01) 15 3.1132) 17.3 142) .42 Male 29.3 (nia) 3021+ 1.0) 30. 5 1+ 0.4) 30. 2 1? 0.5) 32. 0 1+ 1.1pc. 05) Female 7.1 (nia) 7.51+ 0.4) 21+ 0.7) 51+ 0.5) 21+ 0.5) 91+ 0. 5) 2. 2 1p<. 01) Both 15.5 We) 1531+ 0.3) 15.21+0.5) 1751+ 1.4) 1921+ 1.5) 21 g..21+20) 01) 12 +5.7114) +35.3%115) AR Male 25 71nia) 2571+ 0.0) 27. 2 1+ 0.5) 23. 2 1+ 1.0) 31. 7 1+ 3.5) 33. 5 1+ 1. 9) 1.5 5'41 1pc. 05) Female 5.5 (nia) 5.91+ 0.3) 5. 2 1+ 0 4) 91+ 1.7) 7.51- 0.4) 51+ 2.1) 3. 5 31:1 01) Both 10.5 (nia) 11.3 11 .01- 03) 12.01+1.0) 11.31- 0.1) 121 05) 45 1.5145) +14.3%145) CA Male 17.9 (nia) 1341+ 0.5) 17.7-( l0.7) 19.1 1+ 1.4) 13.9 (-0.2) 19.21+0.3) +05% nis Female 4.1 (nia) 501+ 09) 4. 0. 1) 41+ 0.5) 5. 3 1- 0.1) .51+ 0. 3) 7 5'41 1pc. 05) Both 17.31nia) 19.21+1.9) 19. 01-0 2) 20. 0 1+ 1.0) 21.51+1.5) 2321+ 15) 1.3 (psi 01) 3 5.9112) 34.1 122) CO Male 23.5 (nia) 3091+ 2.3) 30.510 .4) 31.51+1.0) 33.41+1.9) 35. 3 1+ 2. 9) +1.4 1p<. 01) Female 7.0 (nia) 3.21+1.3) 321+ 0.0) 11+ 0.9) 10.1 1+ 1.?/61 (cs. 01) Both 9.5 (nia) 3.91? 0.7) 9.1 1+ 0.2) 10. 2 1+ 1.1) 11.01+ 0.3) 1151+ 0.5) +1.5 (cs. 05) 45 1.9143) +192 11211134) CT Male 15.4 (nia) 14.5 1? 1.3) 1501+ 0.4) 15. 5 1+ 1.5) 17. 5 1+ 1.0) 17.31- 0.3) 0.9 his Female 3.5 (nia) 331+ 0.2) 3.71- 0.2) 41+ 0.7) 91+ 0.5) 5.21+ 1.3) 3.5 %1p<.05) Rates are age-adjusted to the US. year 2000 standard. Model-estimated average annual percentage change (MPG) based on all reporting periods; p?yalue indicates statistical signi?cance oftrend; nis indicates trend not significant. Current state rank (59 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. 17 Overall rate change is between the first (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest percentage increase 11} to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically signi?cant difference. if Rate based on 20 suicides. 59 Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank} 1? (State REnk} Both 13.6 1n1a1 12.21.41 1191? 0.31 13.61+ 1.71 14.21+ 0.61 14.41+ 0.21 0.9 nis 42 0.81501 5.9 ?it: 1501 DE Male 23. 0 1n1a1 20.31.71 1991- 0.41 23.1 1+ 3.21 22. 71- 0. 41 2351+ 0.81 0.6 ?fa Female 31ni'a1 010 .21 4.61? 0.41 91+ 0.31 6. 41 1.51 21- 0.21 1.6 nl's Both 91nia1 6.41+ 0.51 6.41? 0.01 31+ 0.81 6?01.71 6.91+ 0.31 0.9 ?fa 51 1.01481 16.1 ?it 1451 DC Male 10.71n1a1 11.11+ 0.41 10.31; 0.81 12.71+ 2.41 10. 012 .61 11. 71+ 1.71 0.3 ?it: nis Female 1.71nia1?FT 2.31+ 0.611?r 331+ 1.01 61- 0.71 3.61+ 1018-0181 3.5 ?fa Both 14.81n1a1 1521+ 0.41 14.91- 0.31 16. 3 1+ 1.41 16. 31?0 .1 16.41+ 0.11 0.8 301p<.051 29 1.61451 10.6 i131481 FL Male 24.3 1n1a1 24.41+ 0.11 23.61- 0.81 26. 2 1+ 2.61 25. 61 0. 61 25.6 1?0. 11 0.5 ?it. Ms Female 6 3 (ma) 681+ 0.51 6.81+ 0.0111+ 0.31 7.61+ 0.51 81+ 0.31 1.4 301p<011 Both 12.91n1a1 13.21+ 0.31 12.31? 0.91 13. 2 1+ 0.91 13.71+ 0.51 1501+ 1.31 0.9 ?it; nits 39 2.1 1401 +162 131441 GA Male 22.1 1n1a1 23.1 1+ 1.01 21311.81 .91+ 0.61 2261+ 0.71 24.-41+ 1.71 0.5 ?fa nl's Female 5.0 1n1a1810.21 4.6 1- 0.2151+ 0.91 5.8 1+ 0.31 6.6 1+ 0.81 2.1 ?fa 1p<.051 Both 12.91nla1 11. 1 11.81 10.31- 0.71 14. 5 1+ 4.11 14.41? 0.11 1521+ 0.81 2.0 ?fa 35 2.41351 18.3 I1111381 HI Male 20.4 1n1a1 17.21.11 15.3 1- 1.91 21. 91+ 6.71 2251+ 0.51 24.31+ 1.81 2.1 ?fa Female 5.4 1n1a1010.41 551+ 0.5111+ 1.51 21- 0 91 5.91- 0.31 1.2 ?fa nl's Both 17.31n1a1 1921+ 2.01 18.31? 0.91 21. 61+ 3.31 2191+ 0.31 24.71+ 2.81 2.3 3611:1101} 6 7.51 61 43.2 1 71 ID Male 28.4 1n1a1 33.1 1+ 4.71 31.1 1? 2.01 3 91+ 3.81 71- 0 21 3801+ 3.31 1.6 301p<051 Female 7.2 1n1a1 6.1 1- 1.11 6.1 1+ 0.0101+ 2.91 9.51+ 0.51 11.81+ 2.31 4.4 1p<.051 Both 9.9 1n1a1 9.81? 0.11 9.71- 0.11 10. 6 1+ 0.81 11.21+ 0.61 1221+ 1.01 1.5 3131p<.051 44 2.31381 22.8 ?it 1321 IL Male 17.1 1n1a1 16.71? 0.41 16.2 1- 0.41 17. 61+ 1.41 18.51+ 0.91 1981+ 1.31 1.1 36113105) Female 3.7 1n1a1 3.814 0.01 381+ 0.2121+ 0.41 4.51+ 0.41 521+ 0.61 2.4 301p<011 Both 13.01n1a1 13.71+ 0.71 14.41+ 0. 71 1491+ 0.51 16.41+1.41 17.11+ 0.71 +1.9 31: 1p<.011 26 4.11231 31.9 13125} IN Male 22.4 1n1a1 23.21+ 0.81 24. 41+ 1.21 24. 71+ 0.41 2671+ 2.01 28.31+ 1.61 1.5 36113101} Female 4.6 1n1a1 501+ 0.41 531+ 0.2191+ 0.61 681+ 0.91 6.61? 0.21 2.7 361131011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia1 is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons (Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 200? 2008 2010 2011 2013 2014 - 2016 Rank 5 (State Rank) 17 (State ank) Both 11.8 (nia) 13.2 1.4) 12 8 0.4) 14.2 1.4) 15.9 16.0 0.1) 2.1 31 4.3 (20) 36.2 ?it: (18) IA Male 20.6 (nia) 22.1 1.5) 20.8 (-1.4) 23.3 2.5) 26.0 25.7 (0 3.) 1.6 ?fa Female (nia) 4.7 1.0) 30(+ .6) 0.2) 0.6) 0.6) 3.8 Both 13.3 (nia) 15.1 15.3( 0.5) 2.4) 19.4 1.6) +2.2 i1110.14.01) 19 +6.0(11) +450 5) KS Male 22.? (nia) 25.0 2.29.1 3.5) 30.? 1.6) 1.9 Female 4.6 (nia) 6.0 1.4) 7(0 3) 4( 0. 3) 6.8 1.4) 8.4 1.6) 3.2 ?fa Both 14.1 (nia) 15.4 1.3) 16. 7 1 .3) 16.2 (.5) 18.2 2.0) 19.3 1.1) +1.9 31: 20 5.2 (16) 36.6 ?it (16) KY Male 25.0 (nia) 26.8 1.9) 28. 3 1.4) 6-2( 1. 0) 30.1 2.9) 31.? 1.6) 1.4 ?fa Female 4.8 (nia) 0 .4) 0. 8) 0.1) 7.1 0.9) 0.6) 3.2 Both 13.1 (nia.4) 13.6 0 3) 14.4 0.8) 17.0 2.5) +1.6 ?it; 27 3.8 29.3 14(26) LA Male 22.9 (nia) 3 0. 6) 22. 4 0.1) 3. 3 0. 8) 23.? 0.5) 27.3 3.6) 1.1 ?fa Female 4.8 (nia) 7( 0.1) 0. 5) 9( 0. 2) 6.1 1.2) 7.5 1.4) 2.8 31;? Both 14.5 (nla) 1310) 18.9 3.5) 18. 0.4) 2.2 51: 21 4.0 (25) 2?.4 ?11} (29) ME Male 25.0 (nia) 22. 9 2. 1) 24.6 1.7) 25. 7 1.1) 31. 5 .4) 29. 8(1 .3) +1.8 ?fa Female 5.3 (nia) 5.3 (- 0.0) 2( 0.1) 0.7) 6( 1.6) 7.9 0.3) 3.1 ?it. Both 10.0 (nia) 10.3 0.3) 10.1 0. 2) 10.2 0.1) 10.? 0.5) 10. 8 0.Male 17.6 (nia) 17.8 0.4.) 18.2 0.5) 18.0 .2) 0.2 nis Female 3.5 (nia) 3.8 0.4) 0.0) 7(02) 4.1 0.4) 4.5 0.4) 1.3 ?fa Both 7.4 (nia9.8 0.4) 10.0 0.3) 2.3 31: 48 2.6 (34 35.3 (20 MA Male 12.1 (nia) 12.8 15.4 2.1) 15200.2) 16.0 0.8) +2.0 Female 3.3 (nia) 90(4 . 4) 1.0) 0.0.2) 3.0 Both 11.8 (nia) 12.5 0.7) 12. 9 0 .4) 1315.6 1.1) +1.9 31: 33 3.9 (26) 32.9 Ml Male 20.0 (nia) 20.9 0.9) 21.6 0. 7) .8 1.3) 23. 9( 1.0) 25.0 1.2) 1.5 ?fa Female 4.4 (nia) 4.8 0.0.9) 2.8 31;? Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change (AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate (1) to lowest rate (51). Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest increase (1) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2016) reporting periods. Ranks are from largest percentage increase (1) to largest percentage decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2015 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period) Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2005 2010 2011 2013 2014 - 2015 Rank 5 {State Rank) 17 (State ank) Both 10.71n1a) 11.51+ 0.9) 12.41+ 0.5) 12.91+ 0.5) 14.21+1.3) 15. 01 0. 9) 2.3 701p<.01) 35 4.3119) 40.5 fit-r1 5) MN Male 15.3 1nia) 1931+ 1.1) 20.41+ 1.0) 2091+ 0.5) 22.9 81+ 1.9) 2331+ 0.4) 1.7 ?fa 1p<.01) Female 3.5 1n1a) 4.21+ 0.5) 51+ 0.5) 1 1+ 0.4) 05) 591+ 1.2) 4.2 55113101) Both 12.91nia) 14.1 1+ 1.2) 14. 7 1+ 0. 5) 1551+ 0.5) 1551+ 0.1) 15. 210 .3) +1.1 3511:1105) 35 2.3135) 17.8 i55140) M5 Male 22.9 1n1a) 24.5 1+ 1.7) 25.1 1+ 0 5) 2551+ 1.7) 25.910 .9) 25. 31.5) 0.7 nlis Female 4.3 1nia) 501+ 0.541+ 0.9) 210.2) 2.4 1p<.01) Both 14.71n1a) 14.1 1? 0 5) 15.41+1.3) 1501+ 0. 7) 17.51+ 1.7) 20. 0 1+ 2.3) 2.2 15 5.3115) 35.4 i15117) MD Male 25.3 1nia) 23.71? 1.5) 25. 5 1+ 1 .9) 2551+ 1.0) 2591+ 2.3) 32. 21+ 3.3) 1.5 501p<.05) Female 5 41n1a) 5.41+ 0.1) 11+ 0. 7) 31+ 0.2) 7.41+ 1.1) 51+ 1.2) 3.2 501p<.01) Both 21.1 1nia) 2251+ 1.4) 23. 5 1+ 1 .0) 24.71+1.1) 2571+ 2.0) 29.21+ 2.5) 2.1 1 5.01 2) 35.0 35111) MT Male 35.9 1n7a) 3731+ 0.0.1) 41 0.1+ 1 .4) 4551+ 4.4) 1.3 55113101) Female 5.71nia) 5.41+1.5) 41 0.1) 10. 0 1+ 1. 5) 12.51+ 2.5) 1311+ 0.5) 4.5 3511:1101) Both 12.71ni'a) 1221? 0 5) 12. 51+ 0 .4) 11. 71?0 5) 1351+ 1.5) 14.51+1.3) +1.0 ?fa 40 2.1 142) 15.2 i111143) NE Male 22.21nia) 2071-1523.91+19.) 0.5 ?fa nis Female 3.5 1n7a) 4.21+ 0.4)11+ 0 9) 01 1.2.5 ?70 his Both 23.31nia) 22.51- 0 5) 22.1 1 0. 5) 2251+ 0. 5) 21.4 2) 23.1 0.2 ?it. n15 9 0.2151) -1.0 55151) NV Male 35.3 1n1a) 35.714 1.7) 35.1 1 1 .5) 35510.7 51; 1115 Female 5.91nia) 951+ 0.5) 51+ 0.1) 100111.21+ 0. 5) +1.5 ?fa 1p<.01) Both 13.51n1a) 12.51?1.) 13.31+0.5) 15.21 1. 9) 15.51+0.5) 20.0 +2.7 3131p<.05) 17 +5.51 5) +453 %1 3) NH Male 22.51n1a) 21.1 11.4) 21.71+ 0. 5) 24. 5 1+ 3.2.2 5511:1105) Female 5.3 1n1a) 510.5) 91+ 1.3.9 501p<.05) Both 7.5 1n1a)7?01 1) 51 0.2 01+ 0.551.5147) 19.2 35135) NJ Male 13.01n1a) 13.11+ 0.0) 12.51 0.55) 13. 71+ 1.1 14.51+ 0.5) 1451+ 0.1) 0.9 %1p<.05) Female 3.2 1nia) 91? 0.3) 301+ 0.0) 91 0.Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nits indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2015. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 2001) and last (2014 2015) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last 12014 2015) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per100,000 Persons {Change from Prior Period}* Current Overall Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2006 2010 2011 - 2013 2014 2016 Rank 5 {State Rank) 17 (State REnk} 55m 22.01n1a1 2201- 011510.21 2301+ 1.21 24.1 1+ 1.11 25.01+1.91 +1.1351p<.051 4 4.01241 15.3 351391 NM Male 36.6 1nfa1 3737.1 1+ 1 .31 4071+ 3.61 0.4 35 nis Female 5.5 1n1a1411.1121011.71+ 0. 91 1201+ 0.31 3.3 351114.051 Both 7.2 1n1a1 110 .1+1. 11 310 .11 21351114011 49 2.1 1411 25.5 351271 NY Male 12.51n1a1 12.210.31 1291+ 0.71 1391+ 1. 01 1541+ 1.41 14.510 .51 1.4 351p<.051 Female 2.71nfa161-0.11 301+ 0311+ 0. 51 4.2 1+0. 71 461+ 0.51 +4.2 35 1p<.011 55m 13.61nl'a1 13510 .11 1371+ 0.11 14.21+0.51 14.5.1+041 15..31+051 +0.5 35154011 34 1.71441 12.7 35 1471 no Male 22.7 1n1a1 2271+ 0.01 22.2 1- 0.51 2331+ 1.11 2331n75 Female 5 61nfa1 51- 0 21 621+ 0.2.0 35113905} Both 13.31nl'a1 14.61+1.31 1601+ 1.41 16. 6 1+ 0. 61 1641+ 1.91 20. 9 1+ 2. 51 2.9 351p<.011 14 7.61 51 57.6 351 11 no Male 21.4 1n1a1 2451+ 3.21 23012.5 351114.011 Female 5.5 1n1a1 51- 1.01 371- 0.5171+ 2 01 5. 7 1+ 1.01 551+ 1 5.1 3.9 35 n75 Both 11.5 1n1a1 1231+ 0.51 131 1+ 0.51 1341+ 0.21 14.51+ 1.41 15.51+1.01 +2.0 35 1p<.011 32 +4.21211 +350 351191 OH Male 20.4 1n1a1 2091+ 0.51 2221+ 1.31 221p<.011 Female 4.0 1n1a1 471+ 0.71 491+ 0.1131+ 0.51091 5. 7 1+ 0. 51 3.4 351114.011 Both 17.0 1n1'a1 16. 511641+ 1.11 20. 7 1+ 2 31 2351+ 2.61 2.3 35109051 7 6.41101 37.6 i151121 OK Male 26.5 1nfa1 27.31.21 2761+ 0. 51 3031+ 2.51 3341+ 3. 11 3731+ 3.61 2.0 351139051 Female 6.6 1nfa1401- .21 7. 5 1+ 1.11 01- 0.51 651+ 1.61 10.31+ 1.61 2.9 1p<.051 Both 16.41nta1 17.71+1.31 17.71.01 16. 61+ 0.91 19. 6 1+ 1 .21 21.1 {+1.31 +1.6 351p<.011 13 4.61161 26.2 i551261 OR Male 27.41nl'a1 2951+ 2.11 25.51 0. 91 29.51+1.01 31.41+1.51 33 01+ 1.51 +1.1 351154.011 Female 6.51nfa1 1 1+ 0.61 7.71+ 0.6141+ 0.71 6612.7 35109011 Both 12.1 We] 1251+ 0.41 12.61+ 0.31 13. 9 1+ 1.11 15. 0 1+1. 11 16.31+1.21 2.0 351p<.011 30 4.1 1221 34.3 351211 PA Male 21. cm n1a a1 2131+ 0.31 21 .91+051 231 1+ 1.21 2471+ 17.1 25.11+1.31 1.5 351114.011 Female 4.2 1nl'a1 451+ 0.31 451+ 0.0141+ 0.91 5. 0 1+ 0.5111+ 1. 11 3.5 35 1114.011 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank (50 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last 12014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 7: 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2016 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period1" Current Overail Overall State Sex [10:35? State Rate Change 2:21:19"; 1999 2001 2002 2004 2005 2007 2008 2010 2011 2013 2014 - 2016 Rank 5 {State Rank) 17 (State REnk} Both 9.41n1a10103101 0.0 01 1281+ 3. 81 11.91.91 12. 6 1+ 0. 71 2.6 ?fa 1p<.051 43 3.2130 W1 34.1 ?fa 123 W1 RI Male 15.41n7a1 1521-0 .21 14.81- 0. 31 21.21+6.41 1912?: 2.01 19..61+041 +2.2 ?fa Female 4.0 1n1a1310.7181+ 0. 41 11+ 1.31001 6.1 1+ 1 .01 3.7 ?fa 1p<.051 Both 12.81n1a1 13.01+ 0.1601+ 1.11 17.71+1.71 2.3 ?fa 1p<.011 23 4.91171 38.3 ?161101 SC Male 21.31n1a1 22.5 1+ 1.21 22.31.11 24. 6 1+ 2. 21 26.11+1.51 28.01+1.91 +1.8 ?fa 1p<.011 Female 5 4 [n7a171-0.7101+ 1.3121+ 0. 21 701+ 0.81 8. 4 1+ 1.41 3.4 ?fa 1p<.051 Both 15.71n1a1 15.81+0. 11 17.1 1+ 1.31 19.31+2.21 1971+ 0.41 22.6 +2.91 +2.5 ?fa 1p<.011 10 +7.01 71 +44.5%1 61 SD Male 27.6 [n1a1 26. 311 .31 27. 9 1+ 1.61 30.1 1+ 2. 21 3201+ 1.91 33. 6 1+ 1.61 1.6 ?fa 1p<.011 Female 4.2 1n1a1 5.81+1.6141+ 0 6131+ 2. 01 31-1 01 11. 31 4. 01 5.8 ?fa 1p<.011 Both 14.61n1a1 1521+ 0.61 16111721+ 0.01 18. 2 1+ 1 .01 +1.4 ?fa 1p<.011 22 3.51281 24.2 ?161311 TN Male 25.1 1n1a1 2541+ 0.31 2681+ 1.31 8 01+ 1. 21 2861+ 0 61 29. 8 1+ 1.21 1.2 ?fa 1p<.011 Female 5.4 [n1a1 631+ 0.91 6 71+ 0 4151+ 0. 81 6.9 (a 0.61 7. 6 1+ 0. 71 1.9 ?fa 1p<.051 Both 12.21nfa1 12.71+ 0.61 12.31- 0.41 13211451+ 0. 91 +1.1 ?fa1p<.011 41 2.31371 18.9 ?161361 TX Male 20.4 [n7a1 2091+ 0.51 20.4 1- 0.61.0123.1 1+ 0. 91 0.9 ?fa 1p<.051 Female 4.8 [his] 5.4 1+ 0.61 5.0 1+ 0.4121+ 0. 21 0.41 6. 4 1+ 0. 81 1.6 ?fa 1p<.051 Both 17.21n1a1 19.0 1+ 1.81 18.21? 0.2521+ 1.21 2.7 ?fa 1p<.011 5 8.01 3 46.5 ?fa 1 4 1W1 UT Male 28.21n1a1 31.1 1+ 2.91 29.4 1- 1.71 32 512.1 ?fa 1p<.051 Female 6.8 [n7a1 7.41+ 0.61 7.51+ 0.2.11 1261+ 2.01 4.4 ?fa 1p<.011 Both 13.21n1a1 16.2 (+3.01 14.91-131 16.61+1.71 18.71+ 2.11 19.71+ 1.01 2.4 ?fa 1p<.011 18 6.41 91 48.6 ?Va 1 21 VT Male 23.6 [n1a1 28. 31+ 61 24.31- 4.01 2731+ 3.01 3101+ 3. 71 3251+ 1.51 1.9 ?fa 1p<.051 Female 4.3 1n1a1 21+4 0. 91 641+ 1.31 6 61+ 0.21 7. 31+ 0. 71 761+ 0.31 3.8 ?fa 1p<.011 Both 12.81n1a1 12.71-011 1291+ 0.31 1361+ 0.71 1461+ 0. 91 1501+ 0.51 +1.2 ?fa 1p<.011 37 2.21391 17.4 ?161411 17.4 Male 21.6 1n1a1 21.31- 1 2101? 0.41 2251+ 1.51 23.61 1 2.1 2391+ 0.21 0.9 ?fa 1p<.051 Female 5.3 [n1a1 5.21? 0.11 591+ 0.71 5.61? 0.31 6. 4 1+ 0 81 691+ 0.51 1.8 ?fa 1p<.051 Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC1 based on all reporting periods; p-yalue indicates statistical signi?cance of trend; indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2016. Ranks are from highest rate 111 to lowest rate 1511. Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1999 20011 and last (2014 2016) reporting periods. Ranks are from largest increase 111 to largest decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 20011 and last (2014 20161 reporting periods. Ranks are from largest percentage increase 111 to largest percentage decrease 1511. Differences between ranks do not necessarily imply a statistically significant difference. TT Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Table 1. Trends in Suicide Rates among Persons 2 10 Years of Age, by State and Sex, National Vital Statistics System, 1999 2018 Age-Adjusted Annual Rate per 100,000 Persons {Change from Prior Period} Current Overall Overall State Sex State Rate Change 2:21:19"; 1999 2001 2002 - 2004 2005 2007 2008 2010 2011 - 2013 2014 2018 Rank 5 {State Rank) 17 (State ank) Both 14.8 (nia) 15411781+ 1.0) 1.1 5511:1405) 24 2.8133) 18.8 c15137) WA Male 24.? [nfa) 2521+ 0 5) 2527.1 1+ 1.1) 13.5 '14 Female 5.0 (We) 8. 4 1+ 0 8) 81+0 .7) 7. 71 0. 8) 851+ 0.8) 2.5 '181p<.01) Both 15.81n1a) 1721+1.8) 1501?0 .7) 19.2.1+32) 21.41+2.2) +18% 11 +58113) +37.1 i?15114) WV Male 27.2 (nia) 3011+ 2. 9) 5.81 1.+11% Female 5.3 (Na) 551+ 0.581+ 2. 2) 3.7 ?it: Both 13.1 1111a) 13.5.1+04) 1501+ 1.0) 15.31+0.3) 1551+ 1.2) 1.5 31: 1p<.01) 28 +3.4129) +258 ?it 130) WI Male 21.7 (We) 22. 2 1+ 0.5) .01+ 1.2) 24. 4 1+ 0 .4) 25.751+ 1.3) +1.1%1ps01) Female 5.1 (nia0.1) 1+ 1.0) 2.5 38113101) Both 20.7 1111a) 23.41.+27) 25.41+28) 28.91+3.5) 28.581- 01) +2.3 251p<.01) 3 +8.1 1 1) 9) WY Male 34.8 (We.2) 47. 1158) 44. 8 1 2. 4) 1.8 ?fa 1p<.05) Female (ma) 521+ 0.5) 41+ 5 52) 1+ 1.4) 1251+ 1. 9) 3.2 '14. 1p<.01) Rates are age-adjusted to the U5. year 2000 standard. Model-estimated average annual percentage change 1AAPC) based on all reporting periods; p?yalue indicates statistical signi?cance of trend; nis indicates trend not significant. Current state rank 150 states and the District of Columbia) is for the reporting period 2014 2018. Ranks are from highest rate 11) to lowest rate 151 Differences between ranks do not necessarily imply a statistically signi?cant difference. ?7 Overall rate change is between the first (1889 2001) and last (2014 2018) reporting periods. Ranks are from largest increase 11) to largest decrease (51). Differences between ranks do not necessarily imply a statistically significant difference. Overall percent change in rates is between the ?rst (1999 2001) and last (2014 2018) reporting periods. Ranks are from largest percentage increase 11) to largest percentage decrease 151). Differences between ranks do not necessarily imply a statistically significant difference. 0 Rate based on 20 suicides. Percentage of injury deaths for which intent was not determined exceeded 20% for both the first and last periods and might have contributed to lower reported rates. Percentage of injury deaths for which intent was not determined declined notably between the first and last periods and might have contributed to the reported rate increase. Figure 1 Percentage Change in Annual Suicide Rates (per 100,000, Age-Adjusted) by State 2014-2016 Compared Against 1999-2001 Increase 37.6% - 57.6% - Increase 31.9% - 37.4% - Increase 18.8% - 29.3% :I ncrease 18.3% :[Decrease 1.0% Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 2? states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted 0R11 {n=20,446} Problemf Mental Square (95% Cl] [95% Cl) (n=9,407} Health Problem (n=11,039] Sex Male 15,702l768} 9,233l83.6} Female 1,806l16.4] p<.01 DANA-0.5) Age" 10?24 1,593l14.4} p<.01 2544 45-64 3,898853] p<.01 65+ p<.01 White, non?Hispanic [0-ch Black, non-Hispanic Mil-4.4] 81717.4) p<.01 American Indian/Alaska Native, non- Hispanic 378(18} 112(12] 2669.4) Asian, non-Hispanic 576(28} 235(25} 3418.1) p<.05 Hispanic ?63019] 6336.7) Other 2103.2) 4s{0.4) Extended demographics Ever served in militarWr 2,075i20.l] Homeless 240(12} 104(1.1] 13641.3) incident Type Single suicide 20,063i982} 10343914) Homicide followed by suicide 319(1.6) 64(0.7} 255(23) p<.01 Multiple suicides 6483.3) 25(03} Method Firearm Ha 23671269] p<.01 Poisoning Substance class causing death?? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marij ua na Tested Positive Antidepressants Tested Positive 1.021(340) 944(314) 300(266) 6241208} 219013) 13,317l65.1} 3,554i41s) 499(6.3} 73619.7) 666(35.8} 603(322) 6441346} 468i25.1} 1951105) 230(33) 6,658i70.8) 4,258i453) 1,238i29.1) 3,866i41.1) 216(56) 376(102} 210(227) 355(311) 336(29.4) 156{13.7) 24(2.1) 315(14) 5.541(50.2) 3.398(353) 325(212) 4.1121372) 283(63) 3.919(355] 360(92) 3.442(312) 261(221) 2.322(210) 479{20.6) p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.Dl p<.01 05004-05) O.5i0.4-0.6] 0.210.103) 0.910.310) 1.111.012) 03(07-03) 1.211.015) 10(09?11) 09(0340) news-.99) 0.5035406) 0.940.910) 09400-10) Alaska, Arizona, Colorado, Connecticut, Georgia, Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carol ina, Ohio, Oklahoma. Oregon. Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. ?Ddds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. Dcecdents were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged 13 years of age and older with reported military service status. Denominator is decedents who died by poisoning, including overdose. Denominator is decedents with any toxicology tested. Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted DR11 Problem* Mental Square [95% Cl] (95% Cl] Health Problem Suicide with known circumstances 18364913) 9,407i100) 9,357l84.8] p<.01 Mental Health Any Current Diagnosed Mental Health Problem? Anxiety disorder Bipolar disorder Schizophrenia 5096.4] PTSD 424{4.5l 226{2.4j Unknown 760{8.1] Current depressed mood p<.01 Substance Problems Any Current substance problem 5319(283) p<.01 Alcohol problem 3,268l17.4) 1,406l15.0] p<.01 Other substance problem 3,084l16.4) p101 Treatment Current mental abuse treatment 5,141l27.4) 64(0.7) p<.01 0.01} Ever treated for mental health/substance 0.02(0.02- problem 394(4.2} p<.01 0.02} Relationship Problems?oss Any relationship problem?oss 7,948t42.4) 3326896) 4,222l45.1) p<.01 Intimate partner problem 5,098l27.2) 2,828l30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131{1.4) 283(30} p<.01 Victim of interpersonal violence within past month 84(0.4) 53(0.6) 31(03) p<.05 Familyr relationship problem 8738.3) 798(35} Other relationship problem (non-intimate) 40342.1) 202l2.1) 201(2.1) Argument or conflict (not specified) 2,914l15.5) 1,636l17.5) p<.01 Death of a loved one (any) 826l8.8) 671(12} p<.01 Non?suicide death 534(57): p<.01 Suicide of family or friend 217(23) 162(1.7) p<.01 Other Life Stressors Any life stressor 9343619) 4,6?5l?93) Recent criminal legal problem 1,588l8.5) 586l6.2) p<.01 Other legal problem 748(4?) 373{4.0) Physical health problem 4.179(223) 2,167l23.2) p<.01 Jobg?Financial problem? 2941(16.2) 1530(16.8) 1411(15.6) p<.05 Eviction or loss of home 7228.8) 405(43} p<.01 School problem?? 162(193) 92(21.9) Recent release from an institution? 941(10.2) p<.01 0.5(0.4u0.5} Jailfprisonldetention facility 203(14.4) 121(25.7) Hospital 51786.6) 31183.0) 206043.?) p<.01 1.3i1.0-1.7) hospital/institution 46983.2) 4i39(46.7) 30(6.4) p?.01 Other (includes alc/SA treatment facilities) 223(15.8) 109(11.6) 114(24.2] p<.01 Recent or Impending Crisis Crisis within past or upcoming two 5,525l29.4) 2,444l26.0) 3,081l32.9) p<.01 Intimate partner problem crisis 1968856) 854(343) 1114(36.2) Physical health problem crisis 739(13.4) 315(12.9) 424(13.8] 1.003.842) Criminal legal problem crisis 621(11.2) 2031283) 418(13.6) p<.01 1.711.511} Family relationship problem crisis 4300.8) 2180.1) p105 Job problem crisis 354(6.4) 1910.8) 163(5.3) p<.01 Suicide Event/History Left a note assatsas) 3,182l33.8) 3,286t35.1) Disclosed suicide intent 4,405l23.5) 2,306l245) 2,099l22.4) p<.01 History of ideation 5,990l313) 3,838l40.8) 2,152l23.0) p<.01 History of attempts 3,732l19.9) 2,770l29.4) 962(103} [3101 *Alaska, Arizona. Colorado, Connecticut. (Lienrgia. Hawaii, Kansas, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey. New Mexico, New York, North Carolina. Ohio, Oklahoma. Oregon, Rhode Island, South Carolina. Utah, Vermont, Virginia, and Wisconsin. Decedent had been identi?ed as having a current diagnosis ofmental health problem in coronen?medical examiner or law enforcement reports. iiOdds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. includes decedents with one or more diagnosed current mental health problems. which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number of decedents with one or more current diagnosed mental health problems. Denominator is decadents aged 18 years of age and older. Denominator is decedents aged 10-18 years. '"15 Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. int? Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. April 10, 2018 Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted (n=20,446) Problem'r Mental Health Square (95% CI) (95% Problem (n=11,039) Sex Male 15,702l76.8) p<.01 Female 4344032} 2338(312) 1,806l16.4) p<.01 Age? 10-24 p<.01 25-44 p<.05 45?64 7,718l37.7) p<.01 Dams-0.8) 65+ 1340(142) p<.01 Race/ethnicity White, noneHispanic p<.01 Black, non?Hispanic 411(4.4) 817024) p<.01 American Indian/'Alaska Native, non?Hispa 378(1.8) 112(1.2) 2669.4) p<.01 Asian, non-Hispanic 5760.8) 235(25) 3418.1) p<.05 Hispanic 19966.4) 463MB) 6336.7) p<.05 Other 66(03} 21(0.2) 45(o.4) p<.05 Extended demographics Ever served in military? 1354(153) p<.01 Homeless 240(1.2) 104(1.1) 136(1.3) 1.103.945} Incident Type Single suicide 93184991) p<.01 Homicide followed by suicide 319(1.6) 255(23) p<.01 Multiple suicides 64(03) 25(0.3) Method Firearm p<.01 1.3(1.7e1.9) 5,907l28.9) 2340813) p<.01 Poisoning p<.01 0-5i0-4-0-5) Substance class causing death? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for a 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive 1.021040) 94401.4) 80006.6) 62400.8) 2190.3) 1.59508) 13.317051) 9.913044) 10.950(53.6) 4.442(40.6) 8.554(418) 2.279066) 8.124097) 2.464003) 7.978090) 4990.3) 7.615072) 7360.7) 6.569(321) 5.425065) 2.214008) 66605.8) 60802.7) 64404.6) 46805.1) 19500.5) 7800.3) 6.653(70.8) 5.192080) 5.409(575) 2.115(391) 4.258(453) 1.238(29.1) 4.226(440) 1.639088) 3.866(41.1) 216(5.6) 3.696(393) 376(10.2) 3.127(33.2) 710(22.7) 3.103030) 1.735050) 35501.1) 33609.4) 156(13.7) 15603.7) 240.1) 8150.4) 6.659(603) 4.721009) 5.541(502) 4.296089) 1.041042) 3.898053) 82501.2) 4.112072) 283(6.9) 3.919055) 360(9.2) 3.442(312) 751(22.1) 2.322010) 47900.6) p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 0.80.709) 0.90.700) 0.30.204) 0.50.406) 0.20.103) 0.90.810) 0.60.607) 0.70.607) 0.80.708) 0.80.709) 0.70.607) 0.40.405) 0.90.809) 1.20.015) 0.90.809) 0.90.800) 0.50.506) 0.20.202) 0.90.7-10) 0.30.304) 0.50.406) 0.20.103) 0.90.810) 0.70.607) 0.80.708) 0.80.809) 0.9(03510) 0.70.708) 0.50.506) 0909-10) 1200-15) 1008-11) 0.60.607) 0.20.203) Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronerr'niedical examiner or law enforcement reports. *Udds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged [8 years of age and older with reported military service status. ?Denominator is decedents who died by poisoning, including overdose. ?Denominator is decedents with any toxicology tested. ?Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- DR5 Adjusted Problem'r Mental Square (95% [95% Cl) Hea?h Problem Suicide with known circumstances 9,407l100) p<.01 Mental Health Am; Current Diagnosed Mental Health Problem? Anxiety disorder Bipolar disorder Schizophrenia 509(5.4) PTSD 424(45) 226(24) Unknown 760(8.1) Current depressed mood p<.01 Substance Problems Any Current substance problem 5,319l283) p<.01 Alcohol problem 3,268l17.4) p<.01 Other substance problem p<.01 Treatment Current mental health/substance abuse treatment 64(0.7) p<.01 {101(00120'01) 0.01l0.01?0.01) Ever treated for mental health/substance problem 6,717l35.3) 394(42) p<.01 032(0322003) Relationship Problems/Loss Any relationship problem/loss 7348(42.4) p<.01 Intimate partner problem 2,270l24.1) 2,828i30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131(1.4) 283(3.0) p<.01 Victim of interpersonal violence within past month 53(0.6) 31(03) p<.05 Family relationship problem 798(8.5) Other relationship problem (non-intimate) 4039.1) 202(2.1) 201(2.1) Argument or conflict [not specified) 2,914l15.5) 1,278l13.6) 1,636l17.5) p<.01 Death of a loved one (any) 826(8.8) 671(72) p<.01 Nonesuicide death 647(5.9} 534(5.7) p<.01 Suicide of family or friend Other Life Stressors Any life stressor Recent criminal legal problem Other legal problem Physical health problem Jobg?Financial problem? Eviction or loss of home School problem?' Recent release from an institution? Jail/prison/detention facility Hospital hospitalfinstitution Other (includes ale/SA treatment facilities) Recent or Impending Crisis Crisis within past or upcoming two intimate partner problem crisis Physical health problem crisis Criminal legal problem crisis Family relationship problem crisis Job problem crisis Suicide Event/History Left a note Disclosed suicide intent History of ideation History of attempts 9,743l51.9) 1588(85) 74801.0) 4,179l22.3) 2941(152) T2288) 162(19.9) 1.412(7.6) 203(14.4) 51786.6) 469(332) 223(1s.a) 5,525t29.4) 1968(35.6) 739(134) 621(11.2) 430(7.s) 354(5.4) 6,468l345) 4.405(235) 5390813) 3,732i1ss) 217(23) 4,675l49.7) 586(62) 378MB) 2,012l21.4) 1530(153) 317(34) 7o(17.a) 941(102} 82(8.7) 31183.0} 439(46.7) 109(11.6) 85484.9) 315(12.s) 203(8.3) 212(s.7) 191(7.a) 2.305(245) 2370994) 152(1?) 5,068l54.2) 1,002l10.7l 370(4.0) 1411(1ss) 405(43) 92(21.9) 471(5.1) 121(25.7) 206(43.7l 30(6.4) 114(242) 3.081(323) 1114(36.2) 424(138) 413(136) 218(7.1) 163(53) 2.152(23o) 962(10.3) p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 ones-es) 1.3(03313) 1.4(1315) 11.58.245.11) roles?1.2) cams-0.9) *Alaska, Arizona, Colorado, Connecticut. Georgia, Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South lCarolina, Utah, Vermont. Virginia. and Wisconsin. 'l Decedent had been identified as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. ?Odds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% (le after controlling for age. sex, race and ethnicity. Known MHP was used as the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number ot'decedents with one or more current diagnosed mental health problems. ll Denominator is decedents aged 18 years of age and older. is aged 10-18 years. 1" Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted (n=20,446) Problem'r Mental Health Square (95% CI) (95% Problem (n=11,039) Sex Male 13702063) 6,469l68.8) p<.01 Female 4344032) 2338812) 1,806l16.4) p<.01 Age?: 10-24 1,211l12.9) 1,593i14.4) p<.01 25?44 3,036l323) 3,420l31?) p<.05 45-54 3,820i40.6) 3,898l353) p<.01 65+ p<.01 Race/ethnicity White, non-Hispanic 8,165l86.8) 8,937l81.0) p<.01 Black, non-Hispanic 411(e.4) 3170.4) p<.01 American lndian/Alaska Native, non-Hispa 378(1.8) 112(12) 2669.4) p<.01 Asian, non?Hispanic 576(2.8) 235(2.5) 341(31) p<.05 Hispanic 453(43) 633(5.7) p<.05 Other 66(03) 21(0.2) 45(0.4) p<.05 Extended demographics Ever served in military? p<.01 Homeless 240(1.2) 104(1.1) 135(1.3) Luce-1.5) Incident Type Single suicide p<.01 news-0.5) Homicide followed by suicide 319(1.6} 255(2.3) p<.01 Multiple suicides 64(03) 25(0.3) 39(0.4) Method Firearm 3,821l40.6) 6,088i553) p<.01 5307983) 236796.51) p<.01 Poisoning 3,003l14.7) 1,861l19.8) p<.01 Substance class causing death?? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for a 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive 1.021040) 94401.4) 80006.6) 62400.8) 2190.3) 1.59508) 13.317051) 9.913044) 10.950(53.6) 4.442(40.6) 8.554(418) 2.279066) 8.124097) 2.464003) 7.978090) 4990.3) 7.615072) 7360.7) 6.569(321) 5.425065) 2.214008) 66605.8) 60802.7) 64404.6) 46805.1) 19500.5) 7800.3) 6.653(70.8) 5.192080) 5.409(575) 2.115(391) 4.258(453) 1.238(29.1) 4.226(440) 1.639088) 3.866(41.1) 216(5.6) 3.696(393) 376(10.2) 3.127(33.2) 710(22.7) 3.103030) 1.735050) 35501.1) 33609.4) 156(13.7) 15603.7) 240.1) 8150.4) 6.659(603) 4.721009) 5.541(502) 4.296089) 1.041042) 3.898053) 82501.2) 4.112072) 283(6.9) 3.919055) 360(9.2) 3.442(312) 751(22.1) 2.322010) 47900.6) p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 0.80.709) 0.90.700) 0.30.204) 0.50.406) 0.20.103) 0.90.810) 0.60.607) 0.70.607) 0.80.708) 0.80.709) 0.70.607) 0.40.405) 0.90.809) 1.20.015) 0.90.809) 0.90.800) 0.50.506) 0.20.202) 0.90.7-10) 0.30.304) 0.50.406) 0.20.103) 0.90.810) 0.70.607) 0.80.708) 0.80.809) 0.9(03510) 0.70.708) 0.50.506) 0909-10) 1200-15) 1008-11) 0.60.607) 0.20.203) Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronerr'niedical examiner or law enforcement reports. *Udds ratio re?ects the risk among those without known mental health problem relative to those with known mental health problems Logistic regression was used to estimate adjusted odds ratio with 95% (313 after controlling for age, sex, race and ethnicity. Known mental health problems was used as the were aged 10 years and older, as per standard in the suicide prevention literature. ll Denominator is decedents aged [8 years of age and older with reported military service status. ?Denominator is decedents who died by poisoning, including overdose. ?Denominator is with any toxicology tested. ?Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- Adjusted Problem1r Mental Square (95% Cl} (95% Cl) Heahh Problem Suicide with known circumstances 18364913) 9,407l100} 9357(843) Mental Health/Substance Problems Any Current Mental Health Diagnosis? 7,076l752) Anxiety disorder 1,579l16.8) Bipolar disorder 1,431l15.2) Schizophrenia 5095.4) 5096.4) PTSD 424(45) 424(45] 226(2.4) 226(2.4) Unknown 750(8.1) Current depressed mood 1038(375) 3.962(421) p<.Ul Substance Problems Any Current substance problem 5,319l28.3) 2376816) 2343030) p<.01 0.740.103) Alcohol problem 3,268l17.4) 1,406l15?) p<.01 Other substance problem 3,084l16.4) 1,758l18.8) 1,316l14.1) p<.01 Treatment Current mental health/substance abuse treatment 3141914) p<.01 Eyer treated for mental health/substance problem 6,323l672) 394(42) p<.01 Relationship Problems/Loss- Any relationship problemlloss 7,948l42.4) 3326896) 4,222l45.1) p<.01 Intimate partner problem 5,098l272) 182880.22) p<.01 Perpetrator of interpersonal violence past month 414(2.2) 131(1.4) 28343.0) Victim of interpersonal violence within past month 31(03) p<.05 Family relationship problem 873(93) Other relationship problem (non?intimate} 403(2.1) 202(2.1) 201(2.1) 1.1(D.9a1.3) Argument or conflict [not specified) 2,914l15.5) 1,278l13.6) 1,636l17.5) p<.01 Death of a loved one (any) 826(8.8] p<.01 Non-suicide death 5345.7) p<.Dl Suicide of family or friend 379(20) 2178.3) 162(1.7) p<.01 Other Life Stressors Any life stressor 9.743819) 4.675(49.7) 5.068842) p<.01 Recent criminal legal problem 1.588(85) 586(62) 1.002(10.7) p<.01 Other legal problem 748(40) 378(4.0) 370(4.0) Physical health problem 4.179823) 2.012814) 2.167(23.2) p<.01 JobXFinancial problem? 2941(16.2) 1530(16.8) 1411(156) p<.05 Eviction or loss of home 7228.8) 3178.4) 405(4.3) p<.01 School problem?? 16289.9) 7087.3) 92(21.9) 1.3(03913) Recent release from an institution? 1.41286) 94180.2) 4718.1) p<.01 0.5(0.4a0.5) Jail/prison/detention facility 203(14.4) 828.7) 12185.7) p<.01 4.58.264) Hospital 51786.6) 31183.0) 206(437) p<.01 hospitalfinstitution 46983.2) 439(467) 30(64) p<.01 Other (includes ale/SA treatment facilities) 223(158) 10981.6) 114(242) p<.01 Recent or Impending Crisis Crisis within past or upcoming two 5.525894) 2.444860) 3.081829) p<.01 1.541.345) intimate partner problem crisis 1968856) 85484.9) 1114862) Physical health problem crisis 739(134) 315(12.9) 42483.8) Criminal legal problem crisis 621(11.2) 2038.3) 418(136) p<.01 Family relationship problem crisis 4308.8) 2128.7) 2188.1) p<.05 Job problem crisis 3548.4) 191(78) 1638.3) p<.01 Suicide Event/History Left a note 6.468845) 3.182838) 3.286831) Disclosed suicide intent 4.405835) 2.306845) 2.099824) p<.01 History of ideation 5.990819) 3.838(40.8) 2.152(230) p<.01 History of attempts 3.732(19.9) 2.770994) 96280.3) p<.01 *Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. New Mexico. New York. North Carolina. Ohio. Oklahoma. Oregon. Rhode Island. South lCarolina. Utah. Vermont. Virginia. and Wisconsin. 'l Decedent had been identified as having a current diagnosis of mental health problem in coroneri?medieal examiner or law enforcement reports. iOdds ratio re?ects the risk among those without known mental health problem relative to those with known mental health problem Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age. sex. race and ethnicity. Known mental health problem was used as the referent Includes decedents with one or more diagnosed current mental health problems. which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number ot'decedents with one or more current diagnosed mental health problems. 9 Denominator is decedents aged 18 years of age and older. is aged 10-18 years. 1" Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted (n=20,446) Problem'r Mental Health Square (95% CI) (95% Problem (n=11,039) Sex Male 15,702l76.8) p<.01 Female 4344032} 2338(312) 1,806l16.4) p<.01 Age? 10-24 p<.01 25-44 p<.05 45?64 7,718l37.7) p<.01 Dams-0.8) 65+ 1340(142) p<.01 Race/ethnicity White, noneHispanic p<.01 Black, non?Hispanic 411(4.4) 817024) p<.01 American Indian/'Alaska Native, non?Hispa 378(1.8) 112(1.2) 2669.4) p<.01 Asian, non-Hispanic 5760.8) 235(25) 3418.1) p<.05 Hispanic 19966.4) 463MB) 6336.7) p<.05 Other 66(03} 21(0.2) 45(o.4) p<.05 Extended demographics Ever served in military? 1354(153) p<.01 Homeless 240(1.2) 104(1.1) 136(1.3) 1.103.945} Incident Type Single suicide 93184991) p<.01 Homicide followed by suicide 319(1.6) 255(23) p<.01 Multiple suicides 64(03) 25(0.3) Method Firearm p<.01 1.3(1.7e1.9) 5,907l28.9) 2340813) p<.01 Poisoning p<.01 0-5i0-4-0-5) Substance class causing death? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for a 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive 1.021040) 94401.4) 80006.6) 62400.8) 2190.3) 1.59508) 13.317051) 9.913044) 10.950(53.6) 4.442(40.6) 8.554(418) 2.279066) 8.124097) 2.464003) 7.978090) 4990.3) 7.615072) 7360.7) 6.569(321) 5.425065) 2.214008) 66605.8) 60802.7) 64404.6) 46805.1) 19500.5) 7800.3) 6.653(70.8) 5.192080) 5.409(575) 2.115(391) 4.258(453) 1.238(29.1) 4.226(440) 1.639088) 3.866(41.1) 216(5.6) 3.696(393) 376(10.2) 3.127(33.2) 710(22.7) 3.103030) 1.735050) 35501.1) 33609.4) 156(13.7) 15603.7) 240.1) 8150.4) 6.659(603) 4.721009) 5.541(502) 4.296089) 1.041042) 3.898053) 82501.2) 4.112072) 283(6.9) 3.919055) 360(9.2) 3.442(312) 751(22.1) 2.322010) 47900.6) p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 0.80.709) 0.90.700) 0.30.204) 0.50.406) 0.20.103) 0.90.810) 0.60.607) 0.70.607) 0.80.708) 0.80.709) 0.70.607) 0.40.405) 0.90.809) 1.20.015) 0.90.809) 0.90.800) 0.50.506) 0.20.202) 0.90.7-10) 0.30.304) 0.50.406) 0.20.103) 0.90.810) 0.70.607) 0.80.708) 0.80.809) 0.9(03510) 0.70.708) 0.50.506) 0909-10) 1200-15) 1008-11) 0.60.607) 0.20.203) Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronerr'niedical examiner or law enforcement reports. *Udds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged [8 years of age and older with reported military service status. ?Denominator is decedents who died by poisoning, including overdose. ?Denominator is decedents with any toxicology tested. ?Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- DR5 Adjusted Problem'r Mental Square (95% [95% Cl) Hea?h Problem Suicide with known circumstances 9,407l100) p<.01 Mental Health/Substance Problems Any Current Mental Health Diagnosis? 107605.21 Anxiety disorder 1,579l16.8) Bipolar disorder Schizophrenia 5095.4) 509(5.4) PTSD 424(45) 424(45) 226(24) 226(24) Unknown 760(8.1) 760(8.1) Current depressed mood 1038815) p<.01 Substance Problems Any Current substance problem 5,319l283) p<.01 Alcohol problem 3,268l17.4) p<.01 Other substance problem p<.01 Treatment Current mental health/substance abuse treatment 64(0.7) p<.01 {101(00120'01) 0.01l0.01?0.01) Ever treated for metnal health/substance problem 6,717l35.3) 394(42) p<.01 032(0322003) Relationship Problems/Loss Any relationship problem/loss 7348(42.4) p<.01 Intimate partner problem 2,270l24.1) 2,828i30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131(1.4) 283(3.0) p<.01 Victim of interpersonal violence within past month 53(0.6) 31(03) p<.05 Family relationship problem 798(8.5) Other relationship problem (non-intimate) 4039.1) 202(2.1) 201(2.1) Argument or conflict [not specified) 2,914l15.5) 1,278l13.6) 1,636l17.5) p<.01 Death of a loved one (any) 826(8.8) 671(72) p<.01 Nonesuicide death 1,181l6.3) 647(5.9] 534(5.7) p<.01 Suicide of family or friend Other Life Stressors Any life stressor Recent criminal legal problem Other legal problem Physical health problem Jobg?Financial problem? Eviction or loss of home School problem?' Recent release from an institution? Jail/prison/detention facility Hospital hospitalfinstitution Other (includes ale/SA treatment facilities) Recent or Impending Crisis Crisis within past or upcoming two intimate partner problem crisis Physical health problem crisis Criminal legal problem crisis Family relationship problem crisis Job problem crisis Suicide Event/History Left a note Disclosed suicide intent History of ideation History of attempts 9,743l51.9) 1588(85) 74801.0) 4,179l22.3) 2941(152) T2288) 162(19.9) 1.412(7.6) 203(14.4) 51786.6) 469(332) 223(1s.a) 5,525t29.4) 1968(35.6) 739(134) 621(11.2) 430(7.s) 354(5.4) 6,468l345) 4.405(235) 5390813) 3,732i1ss) 217(23) 4,675l49.7) 586(62) 378MB) 2,012l21.4) 1530(153) 317(34) 7o(17.a) 941(102} 82(8.7) 31183.0} 439(46.7) 109(11.6) 85484.9) 315(12.s) 203(8.3) 212(s.7) 191(7.a) 2.305(245) 2370994) 152(1?) 5,068l54.2) 1,002l10.7l 370(4.0) 1411(1ss) 405(43) 92(21.9) 471(5.1) 121(25.7) 206(43.7l 30(6.4) 114(242) 3.081(323) 1114(36.2) 424(138) 413(136) 218(7.1) 163(53) 2.152(23o) 962(10.3) p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 ones-es) 1.3(03313) 1.4(1315) 11.58.245.11) roles?1.2) cams-0.9) *Alaska, Arizona, Colorado, Connecticut. Georgia, Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South lCarolina, Utah, Vermont. Virginia. and Wisconsin. 'l Decedent had been identified as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. ?Odds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% (le after controlling for age. sex, race and ethnicity. Known MHP was used as the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number ot'decedents with one or more current diagnosed mental health problems. ll Denominator is decedents aged 18 years of age and older. is aged 10-18 years. 1" Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. Table 2. Select Demographic and Descriptive Characteristics of Suicides among Decedents 310 years of age with and without Known Mental Health Problems--National Violent Death Reporting System, 27 states?, 2015 Characteristics Total Mental Health No Known Chi- Adjusted (n=20,446) Problem'r Mental Health Square (95% CI) (95% Problem (n=11,039) Sex Male 15,702l76.8) p<.01 Female 4344032} 2338(312) 1,806l16.4) p<.01 Age? 10-24 p<.01 25-44 p<.05 45?64 7,718l37.7) p<.01 Dams-0.8) 65+ 1340(142) p<.01 Race/ethnicity White, noneHispanic p<.01 Black, non?Hispanic 411(4.4) 817024) p<.01 American Indian/'Alaska Native, non?Hispa 378(1.8) 112(1.2) 2669.4) p<.01 Asian, non-Hispanic 5760.8) 235(25) 3418.1) p<.05 Hispanic 19966.4) 463MB) 6336.7) p<.05 Other 66(03} 21(0.2) 45(o.4) p<.05 Extended demographics Ever served in military? 1354(153) p<.01 Homeless 240(1.2) 104(1.1) 136(1.3) 1.103.945} Incident Type Single suicide 93184991) p<.01 Homicide followed by suicide 319(1.6) 255(23) p<.01 Multiple suicides 64(03) 25(0.3) Method Firearm p<.01 1.3(1.7e1.9) 5,907l28.9) 2340813) p<.01 Poisoning p<.01 0-5i0-4-0-5) Substance class causing death? Other over-the-counter) Opioids Antidepressants Benzodiazepines Other Toxicology Results Any toxicology testing Positive for a 1 substance? Substance Alcohol Tested Positive Opioids Tested Positive Benzodiazepines Tested Positive Cocaine Tested Positive Amphetamines Tested Positive Marijuana Tested Positive Antidepressants Tested Positive 1.021040) 94401.4) 80006.6) 62400.8) 2190.3) 1.59508) 13.317051) 9.913044) 10.950(53.6) 4.442(40.6) 8.554(418) 2.279066) 8.124097) 2.464003) 7.978090) 4990.3) 7.615072) 7360.7) 6.569(321) 5.425065) 2.214008) 66605.8) 60802.7) 64404.6) 46805.1) 19500.5) 7800.3) 6.653(70.8) 5.192080) 5.409(575) 2.115(391) 4.258(453) 1.238(29.1) 4.226(440) 1.639088) 3.866(41.1) 216(5.6) 3.696(393) 376(10.2) 3.127(33.2) 710(22.7) 3.103030) 1.735050) 35501.1) 33609.4) 156(13.7) 15603.7) 240.1) 8150.4) 6.659(603) 4.721009) 5.541(502) 4.296089) 1.041042) 3.898053) 82501.2) 4.112072) 283(6.9) 3.919055) 360(9.2) 3.442(312) 751(22.1) 2.322010) 47900.6) p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 0.80.709) 0.90.700) 0.30.204) 0.50.406) 0.20.103) 0.90.810) 0.60.607) 0.70.607) 0.80.708) 0.80.709) 0.70.607) 0.40.405) 0.90.809) 1.20.015) 0.90.809) 0.90.800) 0.50.506) 0.20.202) 0.90.7-10) 0.30.304) 0.50.406) 0.20.103) 0.90.810) 0.70.607) 0.80.708) 0.80.809) 0.9(0.85.99) 0.70.708) 0.50.506) 0909-10) 1200-15) 1008-11) 0.60.607) 0.20.203) Alaska. Arizona. Colorado. Connecticut. Georgia. Hawaii. Kansas. Kentucky. Maine. Maryland. Massachusetts. Michigan. Minnesota. New Hampshire. New Jersey. Decedent had been identi?ed as having a current diagnosis of mental health problem in coronerr'niedical examiner or law enforcement reports. *Udds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% CIs after controlling for age, sex, race and ethnicity. Known MHP was used as the reference group. were aged 10 years and older, as per standard in the suicide prevention literature. Denominator is decedents aged [8 years of age and older with reported military service status. ?Denominator is decedents who died by poisoning, including overdose. ?Denominator is decedents with any toxicology tested. ?Denominator for each positive group is the number tested for the substance in that group. Table 3. Circumstances Preceding Suicide among Decedents 310 years of age with and without Known Mental Health Problems--Nationa Violent Death Reporting System, 27 states*, 2015 Characteristics Total Mental Health No Known Chi- DR5 Adjusted Problem'r Mental Square (95% [95% Cl) Hea?h Problem Suicide with known circumstances 9,407l100) p<.01 Mental Health Am; Current Diagnosed Mental Health Problem? Anxiety disorder Bipolar disorder Schizophrenia 509(5.4) PTSD 424(45) 226(24) Unknown 760(8.1) Current depressed mood p<.01 Substance Problems Any Current substance problem 5,319l283) p<.01 Alcohol problem 3,268l17.4) p<.01 Other substance problem p<.01 Treatment Current mental health/substance abuse treatment 64(0.7) p<.01 {101(00120'01) 0.01l0.01?0.01) Ever treated for mental health/substance problem 6,717l35.3) 394(42) p<.01 032(0322003) Relationship Problems/Loss Any relationship problem/loss 7348(42.4) p<.01 Intimate partner problem 2,270l24.1) 2,828i30.2) p<.01 Perpetrator of interpersonal violence past month 414(22) 131(1.4) 283(3.0) p<.01 Victim of interpersonal violence within past month 53(0.6) 31(03) p<.05 Family relationship problem 798(8.5) Other relationship problem (non-intimate) 4039.1) 202(2.1) 201(2.1) Argument or conflict [not specified) 2,914l15.5) 1,278l13.6) 1,636l17.5) p<.01 Death of a loved one (any) 826(8.8) 671(72) p<.01 Nonesuicide death 647(5.9} 534(5.7) p<.01 Suicide of family or friend Other Life Stressors Any life stressor Recent criminal legal problem Other legal problem Physical health problem Jobg?Financial problem? Eviction or loss of home School problem?' Recent release from an institution? Jail/prison/detention facility Hospital hospitalfinstitution Other (includes ale/SA treatment facilities) Recent or Impending Crisis Crisis within past or upcoming two intimate partner problem crisis Physical health problem crisis Criminal legal problem crisis Family relationship problem crisis Job problem crisis Suicide Event/History Left a note Disclosed suicide intent History of ideation History of attempts 9,743l51.9) 1588(85) 74801.0) 4,179l22.3) 2941(152) T2288) 162(19.9) 1.412(7.6) 203(14.4) 51786.6) 469(332) 223(1s.a) 5,525t29.4) 1968(35.6) 739(134) 621(11.2) 430(7.s) 354(5.4) 6,468l345) 4.405(235) 5390813) 3,732i1ss) 217(23) 4,675l49.7) 586(62) 378MB) 2,012l21.4) 1530(153) 317(34) 7o(17.a) 941(102} 82(8.7) 31183.0} 439(46.7) 109(11.6) 85484.9) 315(12.s) 203(8.3) 212(s.7) 191(7.a) 2.305(245) 2370994) 152(1?) 5,068l54.2) 1,002l10.7l 370(4.0) 1411(1ss) 405(43) 92(21.9) 471(5.1) 121(25.7) 206(43.7l 30(6.4) 114(242) 3.081(323) 1114(36.2) 424(138) 413(136) 218(7.1) 163(53) 2.152(23o) 962(10.3) p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.01 p<.05 p<.01 p<.01 p<.01 p<.01 ones-es) 1.3(03313) 1.4(1315) 11.58.245.11) roles?1.2) cams-0.9) *Alaska, Arizona, Colorado, Connecticut. Georgia, Hawaii. Kansas, Kentucky. Maine, Maryland. Massachusetts, Michigan, Minnesota, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South lCarolina, Utah, Vermont. Virginia. and Wisconsin. 'l Decedent had been identified as having a current diagnosis of mental health problem in coronen?medical examiner or law enforcement reports. ?Odds ratio re?ects the risk among those without known mental health problem relative to those with known MHP. Logistic regression was used to estimate adjusted odds ratio with 95% (le after controlling for age. sex, race and ethnicity. Known MHP was used as the reference group. Includes decedents with one or more diagnosed current mental health problems, which are not mutually exclusive. Therefore sums of percentages for the diagnosed conditions exceed 100%. Denominator includes the number ot'decedents with one or more current diagnosed mental health problems. ll Denominator is decedents aged 18 years of age and older. is aged 10-18 years. 1" Denominator of institution subgroup is decedents with recent release from an institution. Recent release from an institution is de?ned as having occurred within the past month. Denominator of crisis subgroup is decedents with any crisis within past or upcoming two weeks. Crises depicted here represent the most commonly occurring categories. 1. All drafts of a document. held by Deborah Stone, that became a report entitled ?Vital Signs: Trends in State Suicide Rates United States. 1999 2016 and Circumstances Contributing to Suicide 27 States. 2015" 2. All drafts of a document, held by Deborah Stone, that became a CDC publication entitled ?Preventing Suicide: A Technical Package of Policy, Programs, and Practices" A Technical Package to Prevent Suicide Prepared by: Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 [Title] is a publication of the National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technical Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. Broadly, the strategies represented include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term effects of suicidal behavior for individuals, families, communities, and society. Specifically, the strategies include strengthening economic supports; strengthening access to mental health care; establishing protective environments; promoting connectedness to protect against suicide; teaching coping and problem-solving skills; identifying and supporting people at?risk; and intervening to lessen harms and prevent future risk. This package supports the National Strategy for Suicide prevention, Goal 1, "Integrate and coordinate suicide prevention activities across multiple sectors and settings.? {p.29} It also supports the National Action Alliance for Suicide Prevention?s priority (2016) "To create and disseminate a framework for comprehensive community-based suicide prevention." Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, businessf?la hot, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome (Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision- making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the CDC, is part of a broader class of behavior called self-directed violence. Self- directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury (Crosby, Ortega, 8n Melanson, 2011). Self-directed violence may be suicidal or non- suicidol in nature. For the purposes of this document, we refer only to behavior where suicide is intended. Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fatal self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicidal behavior presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 in the U.S., the most recent death data available, suicide was responsible for 42,723 deaths, which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2005). In 2014, suicide ranked as the tenth leading cause of death and has been among the top twelve leading causes of death since 1925 in the U.S (Centers for Disease Control and Prevention, 2005). Overall suicide rates have increased from 1999 to 2014 (24% increase) (Curtin, Warner, 8c Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2005). Suicides reflect only a portion of the number of persons affected by suicidal thoughts and behaviors (Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more persons are hospitalized as a result of nonfatal suicidal behaviors than are fatally injured, and an even greater number are either treated in ambulatory settings or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 218 years, for every one suicide there were 9 adults treated in hospital emergency departments for self-inflicted injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide (Ferdon et al., In press). Suicides, attempts, and ideation take an immense emotional, physical, and economic toll on individuals, families and communities. By one estimate, for every death by suicide six people are directly impacted (Le. survivors). Based on this figure it is estimated that there are over 13 million survivors in the US. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8: Sacks, 2002). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. It occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another and act cumulatively to increase one?s vulnerability to think about or engage in suicidal behaviors. The social-ecological model is a useful framework for viewing and understanding suicidal risk factors that have been identified in the literature (Dahlber 8: Krug, 2002): Relationship Individual Risk Factors for Suicide (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012: World Health Organization, 2014) Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, previous victimization, acute and chronic stressors financial problems), genetic and biological determinants, hopelessness Relationship: High conflict or violent relationships, sense of isolation and lack of social support, family history of suicide, financial and work stress Community: Inadequate community connectedness, barriers to health care-- lack of access to providers or medications Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking An individual having or experiencing one or a number of risk factors does not always result in suicide; for example, the vast majority of individuals who are depressed do not die by suicide. It is also important to note that the risk factors described above is not an exhaustive list. These and many other risk factors exist and can be arranged differently or contribute to multiple areas within the social- ecological model. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (1.1.5. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014) Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, peers, and family, connectedness to school, community and other social institutions and the availability of physical and mental health care (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide is connected to other forms of violence in a number of ways. First, suicide and other forms of violence often share some of the same root causes. They can all take place under one roof, or in a given community and can happen at the same time or at different stages of life (Butchart, Phinney, Check, 8: Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, Irwin, Spilsbury, 8t Korbin, 2007; Freisthler, Merritt, 84 LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8L Kim, 2012), intimate partner violence (Pinchevsky Wright, 2012), and youth violence (Sampson, Morenoff, 8t Gannon-Rowley, 2002). Lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoii, Ouyang, 8: Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, Ikeda, Hassan, 8: Ramiro, 2002), intimate partner violence (Heise 81 Garcia-Moreno, 2002; Pinchevsky 8.1 Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Also, while most people who are victims of violence do not act violently or die by suicide, people who experience or are exposed to suicide are at a higher risk for both being a victim of other forms of violence and for inflicting harm on others. For example, children who experience physical abuse or neglect early in their lives are at greater risk for suicide (Briere, Madni, Godbout, 2015), and also at greater risk for committing violence against peers (particularly for boys) (Logan, Leeb, 8: Barker, 2009), bullying (Duke, Pettingell, McMorris, Borowsky, 2010), teen dating violence (Duke et al., 2010), and committing child abuse, elder abuse, intimate partner violence (American Association, 1996), and sexual violence (Jewkes, 2012), later in life. There are also a number of protective factors that pose an opportunity to protect individuals and communities from suicide and other forms of violence, and buffer the effects of shared risk factors. For example, connectedness increases people?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, Choi, 2013; Borowsky, Hogan, 8.: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, 8: Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8a Allen?Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, Vella-Zarb, 2009; Maimon, Browning, Brooks?Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8: Farrington, 2012; Maimon et al., 2010}, and pro-social peers (Capaldi et al., 2012; Losel 8: Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide has far-reaching consequences for individuals, families, and communities. Research indicates that the health consequences of violence, even self-directed, are much more comprehensive than merely injury and death. Suicide attempt survivors may suffer from health consequences ranging from anger, guilt, and mental health problems to traumatic brain injury and physical impairment, depending 9 on the means and severity of the attempt (Chapman 8: Dixon-Gordon, 2007). Further, for each person who dies by suicide, it is estimated that there are an estimated 18 people (called survivors) lCerel, 2015) who experience a major life disruption, such as complicated grief (Mitchell, Kim, Prigerson, 8L Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cvinar, 2005; Runeson 8i Asberg, 2003). The economic toll of suicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8: Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million {Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting {Silverman 8: Maris, 1995; US. Office of the Surgeon General 84 National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family-, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, laborHNational Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014). According to CDC Director, Tom Frieden, successful public health programs also require political commitment, funding, communication, and performance monitoring {Frieden, 2014). Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous leg, RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U5. context if the program, policy, or practice has been evaluated in another country. 10 Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness}. In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training] are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality oftheir implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross?Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below}. The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt]. Preventing Suicide Strategy Approach Strengthen economic supports - Strengthen financial security Housing stabilization policies Strengthen access to mental health care I Coverage of mental health conditions in health 11 insurance policies Establish protective environments Means restriction I Organizational policies and culture I Community?based policies to reduce excessive alcohol use Promote connectedness to protect against 0 Peer norm approaches suiCide I Community engagement activities Teach coping and problem-solving skills - Socialeemotional learning I Parenting skill and family relationship approaches Identify and support people at risk I Gatekeeper training I Screening combined with ca re management I Crisis intervention Intervene to lessen harms and prevent a Treatment for people atvrisk of suicide future risk I Treatment to prevent rerattempts I Postvention I Safe messaging following a suicide The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. [Add text for other issues to potentially cover in this section. You may want to take a look at the other packages in this regard] This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business/labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Research consistently links difficult economic conditions, unemployment, and home foreclosures to increased rates of suicide (Fowler, Gladden, Vagi, Barnes, 8; Frazier, 2015; Luo et al., 2011}. Strengthening financial security may prevent suicide by reducing stress and negative economic outcomes associated with unemployment or underemployment, such as home foreclosures and evictions, which are also associated with suicide risk. Thus, policies that have the potential to improve the socioeconomic conditions of individuals experiencing economic hardship may reduce the risk of suicide. Approaches Economic and housing supports for individuals and families can be strengthened by improving policies through enhancing financial security and stabilizing housing assistance options during times of economic need. I Strengthen household financial security. Research indicates that economic crises are related to suicide rates. Findings from the U.S. show that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old {Fowler et al., 2015; Luo et al., 2011). Policies that support financial security during difficult economic times have been shown to mitigate the risk of economic crises on suicide rates. Unemployment benefit programs provide income protection during periods of unemployment in an effort to prevent or lessen the economic hardship for those experiencingjob loss and enduring unemployment. These benefits may buffer suicide risk during economic crises by increasing financial security and reducing emotional distress among the unemployed. Greater financial assistance through unemployment benefits has been shown to mitigate the risk of unemployment on suicide. I Housing stabilization policies that aim to strengthen housing stability and security may help to buffer the impact of foreclosures and evictions on suicide, as recent research has drawn an association between housing instability and suicidal behavior. Programs that provide affordable housing and other options for homebuyers such as loan modification programs may be used in conjunction with move-out planning and financial counseling services to minimize the impact of foreclosures and evictions on suicide. 13 The National Neighborhood Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration of the neighborhoods and providing affordable housing options for low, moderate, and middle-income home-hovers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Potential Outcomes Reduced suicide rates Lower foreclosure rates I . Lower eviction rates Evidence To date, much research in this area has focused on the association between economic depressions or recessions and suicide rates; however, some studies have demonstrated the impact of public programs on these associations. I Strengthen household financial security. An examination of variations in U.S. unempioyment bene?tprograms across states demonstrated that the impact of unemployment on suicide was offset in those states that provided unemplovment bene?ts greater than the national average across all states (Cvlus, lemour, E: Avendano, 2014}. Another 0.5. study of the link between unemplovment and suicide risk using data on suicides, length of unemployment, and iob losses found that duration, as opposed to merely the loss ofa job, predicted suicide risk (Classen 8L Dunn, 2012}. Together, these results suggest that to help prevent suicide, not onlv Comment IAI: Tom had a concern about this term so we?ve tried to explain it above. Hopefuilv it?s a bit clearer. The interpretation from the studv isn?t the easiest. should state unemployment benefit programs begenerousiin their financial allocations, but also in their duration. I Housing stabilization policies. Although the Nationoi Neighborhood Program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analvsis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particulariv among working-aged adults (Houle Light, 2014]. Another studv of data from 16 0.5. states participating in the National Violent Death Reporting Svstem found that suicides precipitated bv home foreclosures and evictions increased more than 100% from 2005 [before the housing crisis began} to 2010 [after it had peaked; Fowler et al. {2015}}. Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 14 Strengthen Access to Mental Health Care Rationale Mental illness is a risk factor for suicide. Studies suggest that up to 90% of people who die by suicide may have had a mental illness (Cavanagh, Carson, Sharpe, 8t Lawrie, 2003). Research on state?level suicide rates have been found to be correlated with general mental health measures such as depression {ArsenauIt-Lapierre, Kim, 8i. Turecki, 200-4). While most people with mental health problems do not attempt or die by suicide (leson, Gerhard, Huang, at Stroup, 2015) (and most people who attempt suicide do not die by suicide (Suominen et al., 2004)), assuring access to quality mental health care is critical to suicide prevention. Approaches A major approach to strengthening access to mental health care is to have health insurance policies that include coverage for such services. - Coverage of mental health conditions in health insurance policies. Health insurance policies that allow people with mental health problems to access mental health treatment in the same way that they access health care for physical health concerns can increase use of mental health services, help normalize treatment seeking in the population, reduce of mental illnesses like depression and bipolar disorder, and in turn, reduce rates of suicide and suicide attempts. Potential Outcomes I Increased access to mental health services - Decreased of mental illnesses I Decreased rates of suicide attempts I Decreased rates of suicide Evidence Research suggests that policies supporting health insurance coverage of mental healthcare are associated with decreased suicide rates. - Coverage of mental health conditions in health insurance policies. Using data from the National Survey of Drug Use and Health, Harris, Carpenter, and Bao {2006) found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Subsequent research by Lang [2013), suggests that mental health parity laws are associated with an approximate 5% reduction in suicide rates. This reduction (in 29 states) equated to the prevention of 592 suicides per year and a cost savings of 51.3-31.1 million per suicide prevented (Lang, 2013). 15 Establish Protective Environments Rationale Suicide prevention efforts that focus on both the individual and hisfher environment increase the likelihood of lives saved. Establishing protective environments helps ensure that all of the places where individuals live, work, and play are supportive. Limiting access to lethal means, be it at home or in nature, and particularly when an individual may be most vulnerable, can literally make the difference between life and death. Likewise, creating a work environment conducive to prevention and focused on employee well-being supports the large majority of the population where they spend much of their day. Finally, policies that reduce the availability of alcohol, a potent suicide risk factor, serve to further support individuals and protect the environment in which they live. Approaches The current evidence suggests three approaches with promise for creating environments that protect against suicide. 0 Means Restriction. Modifying the environment to decrease access to lethal means is an important public health strategy for preventing suicide. Acute suicidal crises are often brief and impulsive. Previous research indicates that the interval between thinking about suicide and attempting can be as short as 5-10 minutes {Deisenhammer et al., 2009; Simon et al., 2001}. Getting past the impulse by making it more difficult to access lethal means can be lifesaving. Highly lethal means such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and have high fatality rates about 85% of people who use a firearm in an attempt die from their injury). Research also indicates that most people tend not to substitute a different method when a highly lethal method is unavailable or dif?cult to access (Hawton, 2007; Yip et al., 2012). Removing or restricting access to lethal means changes the context of the potential suicide and whether the outcome will be fatal or non-fatal {Yip et al., 2012). intervening or Suicide Hotspots. These interventions are focused on preventing suicides at locations which offer direct means for suicide or a secluded place that prevents intervention. Suicide hotspots include tall structures bridges and cliffs}, railway tracks, and isolated locations that are popular destinations for suicide (for example, parks). Interventions include barriers to preventjumping and signs and telephones to encourage suicidal individuals to seek help (Cox et al., 2013}. :22: Safe Storage Practices for medications, firearms, and other household products can reduce the risk for suicide by preventing impulsive action and separating individuals from easy access to lethal means. Safe storage practices include education and 16 counseling around storing firearms locked in a secure place leg, in a gun safe or lock box), preferably unloaded and separate from the ammunition. Keeping medicines in a locked cabinet or secure location can also prevent their misuse by children and adolescents (Rowhani-Rahbar, Simonetti, 8: Rivara, 2016; C. W. Runyan et al., 2015). Organizational policies and culture that focus on prosocial behavior, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and that have leadership support from the top down can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation}. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components of Togetherfor Life were designed to foster an organizational culture that promoted mutual support and solidarity among members of the Force, help for problems related to suicide, training of supervisors, managers and all units to improve competencies in identifying suicidal risk and in using existing resources, and an education campaign to improve awareness and help-seeking (Mishara Martin, 2012). The United States Air Force Suicide Prevention Program (AFSPP) serves as an example of an organizational policy inclusive of 11 policy and education initiatives that was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service-wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, Caine, 2003). Community-based policies to reduce excessive alcohol use. Acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts lCherpitel, Borges, Wilcox, 2004). While various community policies exist to reduce excessive alcohol use le.g., zoning limits related to alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age), previous research indicates that policies related to outlet locations and density are more strongly associated with suicide, making these particular policies an important approach to preventing suicide. 17 Potential Outcomes I Increase in safe storage of lethal means 0 Reduction in suicide attempts 0 Reduction in suicide deaths I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016). Means restriction. A meta-analysis of suicide hotspot interventions implemented in combination or in isolation in the US. and abroad found that they reduced suicide (Cox et al., 2013; Pirkis et al., 2015]. For example, the suicide rate from jumping from the Jacques-Cartier bridge in Canada decreased significantly from 0.324 to 0.079 per 100,000 after the installation of a bridge barrier, or from about 10 suicide deaths to 2.5 per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013}. Importantly, the reduction in suicides from the bridge was sustained when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013}. In contrast, a study of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand demonstrated the negative impact of removing suicide hotspot interventions on preventing suicide (Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009}. The removal of the Grafton Bridge barriers resulted in an immediate, substantial, and statistically significant increase in suicide by jumping from the bridge. Both the suicide numbers and rate saw a fivefold increase after the removal of the bridge barriers (Beautrais, 2001; Beautrais et al., 2009}. Another form of means restriction involves implementation of safe storage practices. In a case-control study of firearm-related events identified by medical examiner and coroner offices from 3? counties in Washington, Oregon, and Missouri, and 5 trauma centers, Grossman et al. (2005) found that safe storage practices storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device) were protective for suicide attempts among adolescents. A recent systematic review of clinic and community-based education and counseling around safe storage of firearms found that the provision of safety devices significantly increases safe firearm storage practices compared to counseling alone or providing economic incentives to acquire safety devices (Rowhani-Rahbar et al., 2016). The Emergency Department Counseling an Access to Lethai Means (ED CALM) program trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post study, (C. W. Runyan et al., 2016} found that among the parents contacted at follow-up, 76% reported all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among 18 parents who indicated the presence of guns in the home at the time of the child?s initial emergency department visit, all reported guns were currently locked, compared to 67% reporting this at the time of the initial visit (C. W. Runyan et al., 2016). Organizational policies and culture. After implementation of the Together for Life program, police suicides were tracked over 12 years and compared to rates in the control city of Quebec. Pre-post assessments of learning, interviews, and focus groups were also included. The suicide rate in the intervention group decreased significantly by 78.9% to 6.42 per 100,000 per year compared to 29.0 per 100,000 in the control city (Mishara 8: Martin, 2012). Additionally, using a time-series design to examine the impact of the AFSPP program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003). The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch ofthe program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community-based policies to reduce excessive alcohol use. Several studies on alcohol outlet density suggest that measures to reduce alcohol outlet density can potentially reduce alcohol- involved suicides. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo Ortiz, 2002; Giesbrecht et al., 2015). For example, Giesbrecht et al. (2015) found that both on and off- premises alcohol outlets restaurants where alcohol is served and stores where alcohol is available for purchase to go) were positively associated with alcohol-related suicides in 14 U.S. states, particularly among men (AOR 1.08, and American Indiaanlaska Natives (A013: 1.36; CI: Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8: Remer, 2009). 19 Promote Connectedness to Protect Against Suicide Rationale The quantity and quality of our social connection with others has been linked with suicide dating as far back to Durkheim, who first posited that weak social bonds are among the chief causes for suicidality (Durkheim, 1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009), and social connections can be formed within and between multiple levels of the social ecology (Dahlber 8t Krug, 2002) for instance between individuals peers, neighbors, co-workers}, families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community/ neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009) Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize helpuseeking, encourage reaching out and talking to trusted adults, and promote supporting peers through building connectedness. These approaches are typically delivered in school settings but can also be implemented in community settings. 0 Sources of Strength is a suicide prevention program for adolescents (ages 13-17) and young adults (ages 18-25} that uses peer social networks to enhance protective factors and change unhealthy norms and behavior. The program trains young people to serve as peer leaders and connects them with adult advisors at school and in the community. 20 Peer leaders are taught to use their leadership qualities and social influence to promote strength-based messages intended to change peer group norms around coping practices and problem behaviors self-harm, substance use, and unhealthy sexual practices). The goal ofthe program is to reduce the acceptability of suicide as a response to distress, increase the acceptability of seeking help, improve communication between youth and adults, and to develop healthy coping attitudes and behaviors among youth {Wymam 2014). Community engagement activities. Community engagement is an aspect of social capital and involves residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. 0 Greening vocont urbon spaces. Many cities across the nation have experienced urban abandonment over time or in response to the Great Recession {20021-2009}. Some of these communities have sought to engage community members in the cleaning and greening of vacant lots. These activities bring community residents together to clean and beautify vacant areas sometimes building playgrounds and walking areas and other times merely cleaning up litter and mowing the lawns of abandoned homes. Such activities foster community engagement, prosocial norms, and social cohesion, protective factors for suicide {Branas et al., 2011). Potential Outcomes Evidence Reduction in maladaptive coping attitudes and behaviors Increase in healthy coping attitudes and behaviors Increase in referrals for youth in distressed Increase help-seeking behaviors Positive perception of adult support Current evidence provides some support indicating that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools 21 (6 metropolitan, 12 rural), Wyman et al. {2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010). Community engagement activities. One vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots in 4 areas of Philadelphia, PA, resulting in significant reductions in community residents? self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism (Branas et al., 2011). 22 Teach Coping and Problem-Solving Skills Rationale Building life skills prepare individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem- solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014). Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1936), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters {Pollock 31 Williams, 2004). Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, Bunney, 2002} and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, St Beck, 2012; Townsend et al., 2001) appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use) associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). The Youth Aware of Mental Health Program teaches youth about the risk and protective factors associated with suicide (including knowledge about depression and anxiety}, and helps enhance their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions with interactive workshops combined with a booklet that students take home, educational 23 posters displayed in classroom, and interactive iectures about mental health at the beginning and end of the program (Wasserman et al., 2014). 0 Signs of Suicide {505) is a school-based prevention program for students ages 13-17. The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help-seeking behavior (Schilling, Aseltine, 8: James, 2016). The Good Behavior Game (636) is a classroom?based program for elementary school children ages 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the (386 is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008]. Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8: Hunter, 2010). The incredible Years is a comprehensive group training program for parents, teachers and children is designed to prevent and treat behavioral and emotional problems in children ages 2-12. The program includes 9- 20 sessions (depending on the age of the child) offered in community based settings religious, recreation centers, mental health treatment centers, hospital medical centers}. its goal is to reduce conduct and substance abuse problems, two important suicide risk factors in youth, by improving protective factors such as responsive and positive parent?teacher-child interactions and relationships, emotion self-regulation and social competence (all protective factors for suicide) (Herman et al., 2011). 0 Strengthening 10-14 years is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths' interpersonal and problem-solving skills; and creating family activities to build cohesion 24 and positive parent-child interactions. The premise of the program is that developing of these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guyll, Day, 2002l Potential Outcomes 0 Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) I Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills I Enhance problem-solving and conflict management skills Evidence There are a several programs with evidence that supports teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. In a cluster-randomized controlled trial of mm conducted across 10 European Union countries and 168 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt (OR 0.45, 95%Cl 0.24?0.35; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=0.025) at the 12- month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% {Wasserman et al., 2014}. Additionally, in a randomized controlled trial, $05 was shown to reduce self-reported suicide attempts at 3-months post intervention among participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower-risk participants (Schilling et al., 2016). Finally, in an outcome evaluation of the 686, first graders assigned to 686? reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of 636 students, neither suicidal ideation nor 25 suicide attempts were significantly different between (386 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for 636' to be delivered with precision, consistency, and teacher support. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008}. Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. Several studies have demonstrated the effect of The lncreo?ibfe Years program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, 8: Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Stratton, Hammond, 2003; C. Webster-Stratton Hammond, 1997'; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionally, Strengthening Families has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 26 Identify and Support People At-Risk Rationale In order to be successful in decreasing suicidal behavior, attention must be paid to those who are at- risk or vulnerable. These persons experience risk and occurrence of suicidal behavior at higher than average rates. This group requires particular focus on proactive case finding and retention and access to services. These vulnerable or disadvantaged populations include, but are not limited to, individuals living in lower socio-economic status or with a mental health problem; individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain ethnic minority groups. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care are still key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches This document outlines three approaches that focus on identifying and supporting those who are at- risk. I Gatekeeper training is typically implemented in schools and within health care settings and is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Mental Health First Aid (MHFA), designed for the lay public, consists of three weekly sessions of three hours each. The content covers helping people in mental health crises and/or in the early stages of mental health problems. The crisis situations covered included suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior. The mental health problems discussed included depressive, anxiety and disorders. The co-morbidity with substance use disorders is also covered. Participants learn the of these disorders, possible risk factors, where and how to get help and evidence-based effective help (Kitchener 8L lorm, 2004). 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care to assure that people who may be at high-risk of suicide don't ?slip through the cracks?. These approaches typically employ screening for 27 depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow-up. Programs such as these have demonstrated beneficial effects on depression, suicide ideation, and suicide mortality. 0 Henry Ford Perfect Depression Core {Pre-cursor to Zero Suicide). The overall goal of the Henry Ford Perfect Depression Care program was to eliminate suicide. More broadly, the aim of the program was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety. The redesign focused on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims. The program began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006). Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers and/or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means restriction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. 0 Notionoi Suicide Prevention Lifeline This is a nationwide hotline that operates 24/? and is accessible by phone or a web-based chat function. Trained counselors are on-hand to listen, offer fee and confidential emotional support, and provide referrals for mental health services in the local area (Gould, Cross, Pisani, Munfakh, Kleinman, 2013i Applied Suicide intervention Skiils Training (ASST). This a training program for hotline counselors, emergency workers, clergy, caregivers and others in the community. The ASIST model has three phases of caregiving: connecting, understanding and assisting. The training helps participants identify people who are having thoughts of suicide and to recognize their invitation for help (connecting); to listen to the caller?s reasons for dying and living (understanding); and how to conduct a safety assessment, develop a safety plan for the person at risk, and connect the person at risk to community resources {assisting}. The ASIST training program has been field tested in a variety of settings (Gould et al., 2013). 28 Potential Outcomes Reduction in suicide attempts RedUction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I 0 Increased at?risk individuals in treatment I Increased community members trained to identify at?risk individuals I Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained KSubstance Abuse and Mental Health Services Prevention Program. US Department of Health and Human Services: Rockville, MD, 2014l.l-Iowever, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising {Pena 8t Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. II Aid (MHFA), the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, Improved concordance with Administration. Center for Mental Health Services. Report to Congress: Garret Lee Smith Suicide [Comment Need Gatekeeper training. In a randomized controlled trial of 300 participants of Mentoi Heoith First health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health ofthe participants themselves. All results were statistically signi?cant at p<.05. [Kitchener 8: Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed [Kitchener 8: Jorm, 2006). the impact of the Henry Ford Perfect Depression Core {Precursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide Screening combined with care management and overall continuity of care. An examination of between the baseline years prior to the intervention {1999 and 2000} to the intervention years (2002-2009}. During this time period, the suicide rate fell 32% E. Coffey, 2006; C. E. Coffey, Coffey, 34 Ahmedani, 2013}. Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services. the suicide rate increased (M. Coffey, Coffey, EL Ahmedani, 2015]. 29 Crisis intervention. Notionoi Suicide Prevention Lifeiine in an evaluation of the effectiveness of the National Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow- up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. At the same time, researchers found that participants experienced significant decreases in suicidality over the course of the telephone session, and their levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). Applied Suicide intervention Training (ASIST). The ASIST training program has been field tested in a variety of settings. In a national randomized controlled trial, Gould et al. (2013) assessed the impact of the training across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call to the hotline. Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not reseult in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013i 30 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and have had non-fatal suicide attempts or have engaged in non-suicidal self-injury are at increased risk of subsequent suicide- related morbidity and mortality. Risk of suicidality can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide. Exposure to sensationalized or uninformed reporting regarding suicide-related deaths may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Approaches A broad array of approaches to lesson harms and reduce future risk of suicidality among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. I- Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a improving Mood Promoting Access to Collaborative Treatment (IMPACT) aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. IMPACT facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase) by a depression care manager {Hunkeler et al., 2006). 0 Collaborative Assessment and Management ofSaicidality (CAMS) is a therapeutic framework for suicide-specific assessment and treatment of patient?s suicide risk. It is a flexible approach that can be used across treatment settings and clinician theoretical orientations. The clinician and patient work together in an interactive assessment process. The patient is highly engaged in the development of their own treatment plan. Every session of CAMS is collaborative and involves the patient?s input about what is and is not working. Ultimately, this process is designed to enhance the therapeutic alliance and increase treatment motivation in the suicidal patient {Jobes, 2012). 31 Dialectical Behavioral Therapy is a multicomponent therapy for individuals at high risk for suicide who may struggle with impulsivity and emotional regulation. The components of DBT include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. Attachment-Based Family Therapy (ABFT) is a program for adolescents ages 12-18 that is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). Treatment to prevent re?attempts. These follow-up contact approaches use diverse modalities home visits, mail, telephone, e?mail] to engage recent suicide attempt survivors to prevent reattempts. These approaches typically focus on coping and other emotional regulation skills and may include case management home visits to increase adherence to and continuity of care, one-on-one interpersonal therapy andfor group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%-25% reattem pt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) 0 Emergency Department Brief intervention with Fallow-up Visits - A one-hour discharge information session that addressed suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options was combined with nine follow-up contacts over 18-months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months]. Follow-up contacts were either conducted by phone or home visits according to a specific time line for up to 18-months. Active follow-up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of caring and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years}. Cognitive Behavior Therapy for Suicide Prevention uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CBT-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. 32 Postvention approaches are implemented after a suicide has taken place and may include debriefing of survivors (those who have lost a friend, peer, family member, co?worker to suicide}, counseling, andior bereavement support groups. The programs have not typically tested their impact on suicide or suicidal behavior but may reduce risk of guilt, feelings of depression, and complicated grief (Szumilas Kutcher, 2011). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention {no outreach) (Cerel 8; Campbell, 2008). .3. StondBy Response Service is a suicide bereavement support service. The service provides clients with face-to-face outreach and telephone support provided by a professional crisis response team. A site coordinator then develops a customized case management plan, referring clients to other existing community services matched to their needs (Visser, Comans, 8L Scuffham, 2014). Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Media guidelines. Guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotlinel and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion {Niederkrotenthaler 8t Sonneck, 2007}. Potential Outcomes Reduction in mental health-related sequelae Increase HESS Improved coping skills Improved messaging following suicide Reduction in re-attempts 33 Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation ofsuicide-related mortality is a statistically rare event, evaluation of mortality outcomes requires large sample sizes and extended follow-up. Therefore, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at-risk of suicide. Improving Mood Promoting Access to Collaborative Treatment (IMPACT) has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up [Hunkeler et al., 2006; Unutzer et al., 2006} relative to patients who received care as usual. Collaborative Assessment and Management of Suicidality (CAMS) has been tested and supported in correlational studies (Jobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community?based sample of suicidal outpatients. {Comtois et al., 2011). Dialectical Behavioral Therapy. in a randomized controlled trial of women with recent suicidal or self-injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined {Linehan et al., 2005]. Attachment-Based Family Therapy (ABFT). A randomized controlled trial of ABFT (Diamond et al., 2010) found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care (-4.37 vs. -2.34; .001; Additionally, a higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 45.2%; .006; {Diamond at al., 2010). Treatment to prevent re-attempts. 34 Emergency Department Brief Intervention with Follow-up Visits. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 13-months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively; chi2 13.83, 0.001] {Fleischmann et al., 2003}. Active follow-up contact approaches intended to prevent reattempts among patients that have been hospitalized and subsequently discharged for suicide attempts have been found in a meta- analysis conducted by Inagaki et al. (2015lto reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta?analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of postcrisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8; Carter, 2011; Wang et al., 2016i Postvention. In a study by Visser et al. {2014), StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Safe messaging following a suicide. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 31 suicides annually {95% confidence interval: -149 to -13; -2.32, df 54, 0.024) in the Viennese subway system (Niederkrotenthaler 8: Sonneck, 2007) 35 Sector Involvement Public health can piay an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business/la bor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016). The National Electronic Injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.), age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose {Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts It is also important at all levels ilocal, state, and federal} to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk 37 and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly,r over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community?level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion 38 References {See last 7 pages after Appendix?EndNete put them there because it thought the Appendixftable was part of the regular text; Didn?t want to cut and paste table for fear it?d get tie-formatted) 39 Appendix A: Summary of Strategies and Approaches to Prevent uicid Strategy litmustroacI-uf Program, Practice or Policyr Strengthen ?nancial security Best Available Evidence Suicide Suicide Attempts or ldeation Other Risk;f Protective Factors for Suicide Lead Sectors1 1 Comment I?ll help you complete the lead sector column. For the other columns, you just need to insert a check? mark based on the evidence vou describe In the narrative for a particular program or policy. For example, if the evidence shows impact on suicide, then put a check-mark in that column. If the studllr also found effects on risk or protective factors, then put a check-mark in that Strengthen Unemployment benefit programs Government economic (local, state, supports I . - I I FEderal) Housmg stabilizatlon polIcres The National Neighborhood Stabilization Government Pl" am (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parity Lows Healthca re care Government (local, state, Federal) Means restriction lntervening at hot spots Government (local, state, Establish t' :ederall protective ofe storage pro: toss (a tempts} .3 storage overnment . of firearms and (local, state, enwronments . . medication} Federal} Organizational policies and culture Lcolumn as well. Best Availabl Evidence Togetherfor Life Business/Labor US Air Force Suicide Prevention Program (family Government violence) (local, state, Federal) Business/Labor Community?based policies to reduce excessive alcohol use Aicohoi outiet density Government (local, state, Federal) Peer norm approaches Promote connectedness SDWCES of Strength Public Health to protect Social Services against suicide actwatles Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Manta} Heaith Program Public Health Social Services Signs of Suicide Teach co in 3 Good Behavior Game and problem- solving skills Parenting skill and family relationship approaches The incredibie Years Public Health Social Services Strengthening 10-14 Best Availabl Evidence Gatekeeper training Mentai Heaith First Aid Public Health Healthcare Social Services Identify and Screening combined with care management people Henry Ford Perfect Depression Care {Pre- Healthcare cursor to Zero Suicide) Crisis Intervention Nationai Suicide Prevention Lifeiine Public Health Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide improving Mood - Promoting Access to Healthcare Coiiaborative Treatment Social Services Intervene to Ufogcigg?gtrigjggessment and Management lessen harms and We?? Diaiecticai Behaviorai Therapy future risk Attachment-Based Famiiy Therapy Treatment to prevent rerattempts Best Availabl Evidence ED Brief intervention with Follow-up Visits Health ca re Active follow-up con tact approaches EST for Suicide Prevention Postvention Stono?By Response Service Healthcare Safe messaging following a suicide Medio Guidelines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as nongovernmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. REFERENCES American Association. (1996). Violence and thefamiiy: Report of the American Association presidential taskforce on violence and the family: APA. ArsenauIt-Lapierre, (3., Kim, C., 8: Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. doi: 10.1186f1471-244X-4-37 Ba ndura, A. (1986). Socialfoundations of thought and action: A social cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M, 8: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and violent behavior, 14(5), 336-347. Basile, K. C., Hamburger, M. E., Swahn, M. H., St Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journal of Emergency Medicine, 14(4), 329. Beautrais, A. L. {2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi: 10.1080i0004867'010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M, Horwood, L. J., 8: Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi: 10.1080i0004867'0902873214 Beyer, K. M, Layde, P. M, Hamberger, L. K., 81 Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49?64. doi: Blair, J. M, Fowler, K. A., Jack, 5. P., 8! Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. ini Prev, 22 Suppil, i6?I?11. doi: Borowsky, I. W., Hoga n, M., 81 Ireland, M. {1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M, Tam, v. W., Jackson, T. D., 84 Ten Have, T. R. (2011). A difference? in?differences analysis of health, safety, and greening vacant urban space. Am Epidemiai, 174(11), 1296?1306. doi: 10.1093iajeikwr223 Briere, J., Madni, L. A., Godbout, N. (2015). Recent Suicidality in the General Population: Multivariate Association With Childhood Maltreatment and Adult Victimization. .i interpers Violence. doi: 10.1122/08362605155843139 Butchart, A., Phinney, A., Check, P., 8L Villaveces, A. (2004). Preventing Violence: A Guide to implementing the Recommendations of the World Report on Violence and Health. Geneva, Switzerland: World Health Organization. Capaldi, D. M., Knoble, N. B., Shortt, J. W., 81 Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231-280. doi: 10.1891f1946-6560.3.2.231 Carter, M, McGee, R., Taylor, 8., 8: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adolesc, 30(1), 51-62. doi: Cavanagh, J. T., Carson, A. J., Sharpe, M., Lawrie, S. M. (2003). autopsy studies of suicide: a systematic review. Meal, 33(3), 395-405. Centers for Disease Control and Prevention. (2005). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. from Available at: Strategic Direction Full version-a.pdf Centers for Disease Control and Prevention. (2015a). Child Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2015:). Sexual Violence: Risk and Protective Factors. from Available at: Cerel, 1. (2015). We are ail connected in suicidoiogy: The continuum of "survivorship. Paper presented at the Plenary presentation at the 48th annual conference of the American Association of Suicidology Atlanta, GA. Cerel, 1., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. doi: 10.1521fsuli.2008.38.1.30 Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. doi: 10.1521/suli.2007.37.5.543 Cherpitel, C. 1., Borges, G. L. G., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-285. Classen, T. 1., 84 Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi: 10.1002fhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524-1526. doi: 10.1176ips.2006.5?.1o.1524 Coffey, C. E., Coffey, M. 1., 8; Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. doi: 10.1176/appi.PS.640422 Coffey, M, Coffey, C., 84 Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. doi: Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. (3., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963372. doi: 10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M, Spilsbury, J. C., 8: Korbin, J. E. How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negl', 111741142. doi: Cox, (3. R., Owens, 0, Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. doi: Crosby, A. E., Han, B., Ortega, L., Parks, 5. E., 84 Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 55-13), 1-22. Crosby, A. E., Ortega, L., 8.: Melanson, C. (2011). Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Crosby, A. E., 8: Sacks, 1. 1. {2002). Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav, 32(3), 321-328. Curtin, S. C., Warner, M., 8: Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cvinar, 1. G. (2005). Do suicide survivors suffer social stigma: A review of the literature. Perspectives in care, 41(1), 14-21. Cylus, 1., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am Epidemioi, 180(1), 45-52. doi: 10.1093faje/kwu106 Dahlber, L. L., 81 Krug, E. G. (2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, l. A. Mercy, A. B. Zwi, 8: R. Lozano World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, l. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Ann)i 152(2), 311-318. doi: 10.1176/appi.ajp.162.2.311 Diamond, G. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Duke, N. N., Pettingell, S. L., McMorris, B. J., 8: Borowsky, I. W. [2010). Adolescent violence perpetration: associations with multiple types of adverse childhood experiences. Pediatrics, 125(4), e773-786. doi: 10.1542lpeds.2009-0597 Durkheim, E. (1951). Suicide: A study in sociology (JA Spaulding 8: G. Simpson, trans). Glencae, it: Free Press.(0riglnal work published 189?). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 84 iv'ella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 3? countries. 1 Adolesc Health, 45(4), 351?359. doi: Escobedo, L. (5., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accld Anal Prev, 34(5), 683694. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman 1., Reed, G. A., 84 lskander, J. {In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi: 50042~ 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., ?v'agi, K. J., Barnes, J., Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311-316. doi: 10.2105lAl PH.2014.301945 Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology ofchild maltreatment: a review of the literature and directions for future research. Child Moltreat, 11(3), 263280. doi: 10.1172f1077559506289524 Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi: Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, C., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. doi: 10.1111ladd.12762 Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M., 8: Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(5), Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. doi: 10.1521fsu i.2007.37.3.333 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-214. doi: 10.1001fjama.293.6.707 Harris, K. M., Carpenter, C., 8: Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. doi: Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., St Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Jaurnai of 198(4), 309-316. Hawton, K. (2007}. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi: 10.1027f0227-5910.28.sl.4 Heise, L., 8: Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano {Eds}, Worid Report on Vioience and Health (pp. 87-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 3: Webster?Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's (Io?Occurring Internalizing Q, 26(3), 189?201. doi: 10.1037ia0025228 Hong, J. S., Kral, M. J., Espelage, D. L., St Allen?Mea res, P. (2012). The social ecology of adolescent?initiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431?454. doi: 10.1007/510578? 011?02?3?y Houle, J. N., 81 Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073?1079. doi: Hunkeler, E. M, Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259?263. doi: 10.1136fbmj.38683.210255.BE lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a metamanalysis. JAffect Disord, 175, 66?73. doi: 10.1016fj.jad.2014.12.048 Jewkes, R. (2012). Rape perpetration: A review. Pretoria Sexual Vioiehce Research initiative. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence? based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi: 10.1111fj.1943+ 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12}, 2124-2133. doi: Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., IWang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohoi Depend, 95 Suppi 1, 55-528. doi: Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. doi: Kitchener, B. A., 8: Jorm, A. F. (2005). Mental health first aid training: review of evaluation studies. Aust Zi 40(1), 6-8. doi: Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 8: Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi: 10.1027f0227- 5910la000136 Klevens, J., Simon, T. R., 8L Chen, J. (2012}. Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 22110), 1987-2002. doi: 10.1172f0886260511431441 Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ, 322(2428), 1376. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457-2463. doi: 10.21051ajph.2009.159871 Knox, M. 5., Burkha rt, K., Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 7?3-88. doi: 10.1002Jhec.1816 Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752?266. doi: Logan, J. E., Leeb, R. T., 81 Barker, L. E. (2009). Gender?specific mental and behavioral outcomes among physically abused high?risk seventh-grade youths. Public Health Reports, 234?245. Losel, F., at Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58?523. doi: 10.1016fj.amepre.2012.04.029 Luo, F., Florence, C. 5., Quispe?Agnoli, M, Ouyang, L., St Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928?2007. Am Public Health, 101(6), 1139?1146. doi: 10.2105Jajph.2010.300010 Maimon, D., Browning, C. FL, 8L Brooks~Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 302*324. doi: Mishara, B. L., Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention 8.: Suicide, 33(3), 162?168. Mitchell, A. M, Kim, Y., Prigerson, H. G., MortimermStephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 1218. doi: 10.102710227591025.1.12 Motto, J. A., Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828?833. Muennig, P., Cohen, A. K., Palmer, A., 31 Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. doi: National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention. {2016). Action Alliance Priorities. Retrieved July 12, 2016, from Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust ZJ 41. doi: 10.1080l00048670701266680 Olfson, M., Gerhard, T., Huang, C., 5., 81. Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 22(12), 1172-1181. doi: Pena, J. 3., 8t Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614-632. doi: 10.1521/suli.2006.36.6.614 Perron, 5., Burrows, S., Fournier, M., Perron, P. A., Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi: 10.21051ajph.2012.301089 Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Violence Abuse, 13(2), 112-132. doi: 10.1177/15248380124456111 Pirkis, J., Too, L. S., Spittal, M. J., Krysinska, K., Robinson, J., 8: Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. doi: Pollock, L. R., 8: Williams, J. M. {2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, D., Chang, S. 5., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. doi: 2 Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. 0., Ireland, M., Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. 1 Adolesc Health, 35(5), 424 e421- 410. doi: Rowhani?Ra hba r, A, Simonetti, J. A., St Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemlol Rev, 38(1), 111?124. doi: 10.1093/epirev/mxv006 Ru neson, B., 81 Asberg, M. (2003). Family history of suicide among suicide victims. Am 160(8), 1525* 1526. doi: 10.1126/appi.ajp.160.8.1525 Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. Westl Emerg Med, 17(1), 8?14. doi: 10.5811lwestjem.2015.11.28590 Runyan, D., Wattam, C., Ikeda, R., Hassan, F., 81 Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on Violence and Health (pp. 5986). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., Gannon?Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review afSaciology, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 81'. James, A. (2016). The SDS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 12(2), 157-166. doi: Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, G. A., Jr., 81 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., El Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and life? threatening behavior, 25(1), 10-21. Simon, 0. R., Swarm, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., Day, S. X. (2002}. Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. 1 Stud Alcohol, 63(2), 219- 228. Suominen, K., Isomets'a, E., Suokas, J., Haukka, J., Achte, K., St Lonnqvist, J. (2004). Completed Suicide After a Suicide Attempt: A 37-?lear Follow-Up Study. American Journal of 161(3), 562-563. doi: doi:10.1176fappi.ajp.161.3.562 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can 1 Public Health, 102(1), 13-29. Townsend, E., Hawton, K., Altman, D. (3., Arensman, E., Gunnell, 0., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 929-988. U.S. Office of the Surgeon General, St. National Action Alliance for Suicide Prevention. (2012). 20.12 Nationai Strategy for Suicide Prevention: Goais and Objectives for Action. Washington, DC: HHS. Unutzer, J., Tang, L., Dishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journaiof Community 42(1), 19-28. doi: 10.1002ijcop.21586 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 45(1), 106-120. doi: Wasserman, 0., Haven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. doi: Webster-Stratton, C., 8: Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consuit Ciin 65(1), 93?109. Webster?Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Ciin Chiid 30(3), 283602. doi: Webster?Stratton, C. H., Reid, M. J., 81 Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journai of Ciinicai Chiid 8: Adoiescent 40{2), 191?203. Webster?Stratton, C., Jamila Reid, M., Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high?risk schools. Jaurnai of chiid and 49(5), 471488. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2008). Measuring neighborhood connection and the association with violence in young adolescents. JAdoiesc Heaith, 43(5), 482?489. doi: Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., St. Anthony, J. C. (2008). The impact of two universal randomized first? and secondegrade classroom interventions on young adult suicide ideation and attempts. Drug Aicohoi Depend, 95 Suppl 1, 560?23. Wilson, W. J. (2011). When work disappears: The woria' of the new urban poor: ?v'intage. World Health Organization. (2014). Suicide Prevention: A Giobai imperative. Geneva, Switzerland: WHO Press. Wyman, P. A. (2014). Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med, 47(3 Suppl 2), 5251-255 doi: 10.1016fj.amepre.201?i.05.039 Wyman, P. A., Brown, C. H., LoMurray, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Pubiic Heaith, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., St Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. doi: A Technical Package to Prevent Suicide Prepared by: Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 [Title] is a publication of the National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technical Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. Broadly, the strategies represented include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term effects of suicidal behavior for individuals, families, communities, and society. Specifically, the strategies include strengthening economic supports; strengthening access to mental health care; establishing protective environments; promoting connectedness to protect against suicide; teaching coping and problem-solving skills; identifying and supporting people at?risk; and intervening to lessen harms and prevent future risk. This package supports the National Strategy for Suicide prevention, Goal 1, "Integrate and coordinate suicide prevention activities across multiple sectors and settings." {p.29} It also supports the National Action Alliance for Suicide Prevention?s priority (2016) "To create and disseminate a framework for comprehensive community-based suicide prevention." Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome (Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision- making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury (Crosby, Ortega, Melanson, 2011}. Self- directed violence may be suicidal or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended. Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotoi self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 in the U.S., the most recent death data available, suicide was responsible for 42,723 deaths, which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the tenth leading cause of death and has been among the top twelve leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 {Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 {Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behaviors (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide (Le. ideation) (Ferdon et al., in press). Suicides, suicide attempts, and ideation take an immense emotional, physical, and economic toll (see p. 9) on individuals, families and communities. By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8: Sacks, 2002). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlber Krug, 2002): Relationship Individual Some risk Factors for suicide include Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014): 0 Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness 0 Relationship: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community: Inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness It is important to recognize that the vast majority of individuals who are depressed (or who have other risk factors) do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General El. National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014) Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General at National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. As indicated above, suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes {Butchart, Phinney, Check, Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide {Desai, Oausey, 8i. Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, Irwin, Spilsbury, 8: Korbin, 2007; Freisthler, Merritt, 8: LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8: Kim, 2012), intimate partner violence (Pinchevsky 8: Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, 8: Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, Ikeda, Hassan, 8: Ramiro, 2002), intimate partner violence (Heise 8: Garcia-Moreno, 2002; Pinchevsky 8: Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, 8: Weingarden, 2012; Pinchevsky 8: Wright, 2012; Widome, Sieving, Harpin, 8: Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8: Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, 8: Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen- Meares, 2012; Losel 8: Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8: VeIIa-Zarb, 2009; Maimon, Browning, 8: Brooks- Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8: Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel 8: Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne-Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). Research indicates that the health consequences of violence, including suicide, are much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long?term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, 8: Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2008). The economic toll of suicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8i. Silverman, 2016}. The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million {Shepard et al., 2016]. Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (US. Public Health Service, 1999}. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman 8L Maris, 1995; US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare,justice, education, labor) {National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. According Frieden {2014), successful public health programs also require political commitment, funding, communication, and performance monitoring Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous le.g., RCT or quasi-experimental design} evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility ofimplementation in a US. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness}. In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts} provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and 10 family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality oftheir implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below}. The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen financial security Housing stabilization policies Strengthen access to mental health care - Coverage of mental health conditions in health insurance policies Establish protective environments a Means restriction I Organizational policies and culture I Community-based policies to reduce excessive alcohol use Promote connectedness to protect against I Peer norm approaches suicide I Community engagement activities 11 Teach coping and problem-solving skills - Socialaemotional learning 0 Parenting skill and family relationship approaches Identify and support people at risk - Gatekeeper training Screening combined with care management Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re?attempts I Postvention - Safe messaging following a suicide Intervene to lessen harms and prevent future risk The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Suicide ideation, attempts, morbidity and mortality vary by gender, racex?ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business/labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Comment IAI: lthink we need to describe what this approach is more specifically so that we can then show the evidence in the next section. I 1 Strengthen Economic Supports Rationale Public health research suggests that some of the biggest impacts on health come from policies developed to improve socioeconomic conditions iFrieden, 2010). Downturns in the economy and increases in unemployment and home foreclosures are associated with increased rates of suicide (Fowler, Gladden, Vagi, Barnes, 8: Frazier, 2015; Luo et al., 2011}. Policies that strengthen financial security and keep housing stable may help prevent suicide by reducing stress and anxiety and the potential for a crisis situation and at the same time assist people to improve their financial situations. Approaches Economic and housing supports for individuals and families can be strengthened by improving policies that enhance financial security and stabilize housing for people, especially in times of economic need. I Strengthen household financial security.[Findings from the U5. show that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic estpansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015}. Policies that support financial security during difficult economic I Housing stabilization policies aim to keep people in their homes during times of financial Insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential home-buyers such as loan modification programs, move?out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes Reduced suicide rates Lower foreclosure rates Lower eviction rates Reduced emotional distress Evidence There is evidence that policies that strengthen household financial security and that stabilize housing can reduce suicide risk. I Strengthen household financial security. Unemployment benefit pro-grams provide income protection during periods of unemployment in an effort to prevent or lessen the economic 13 hardship for those experiencingjob loss and enduring unemployment. An examination of variations in U5. unemployment bene?t programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average Comment IAI: Torn had a concern about this term so we?ve tried to expiain It above. Hopefully it?s a bit clearer. The interpretation from the study isn?t the IL Leasiest. Comment A z Did the National Neighborhood Stab. Program increase housing options and decreases evictions for people who participated? I think this I I unemployment benefits greater lelus, Glymour, 8i Avendano, 2014). Another U.5. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk (Classen 8L Dunn, 2012}. Together, these results suggest that not only should state unemployment bene?t programs be generougin their financial allocations, but also in their duration. I Housing stabilization policies. The Notional Neighborhood Program provides affordable housing options for low, moderate, and middle-income homebuyers and offers financial assistance to eligible individuals for the purchase of a new home. While it has not been rigorously evaluated for its impact on suicide outcomes, specifically, it addresses foreclosure and swarms: risk factors for News decreases as assesses- . foreclosures demonstrated that as the proportion of foreclosed properties increased in LLS. states, so did the state suicide rate, particularly among working-aged adults (Houle 8: Light, 2014]. Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 [before the housing crisis began} to 2010 [after it had peaked; Fowler et al. {2015]}. Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move~out planning and counseling services may help to prevent suicide. Lwould be useful to say. ., 14 Strengthen Access to Mental Health Care Rationale Mental illness is a risk factor for suicide-- Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths (Cavanagh, Carson, Sharpe, El Lawrie, 2003} and research on state-level suicide rates have been found to be correlated with general mental health measures such as depression (Arsenault-Lapierre, Kim, Turecki, 2004}. While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, 8: Stroup, 2015) and most people who have attempted suicide will not go on to die by suicide (Owens, 2002}, assuring access to affordable mental health care for people in need is critical to suicide prevention. Approaches A major approach to strengthening access to mental health care is through the provision of health insurance policies that include coverage for such services. - Coverage of mental health conditions in health insurance policies. Historically, mental health care was viewed as separate from physical health care and many health insurance plans either did not provide coverage of mental health services or such services were available but not affordable-due to higher co-pays or co-insura nce?- or set limits on the number of visits allowed. More recently, improvements have been made with many health insurance policies providing greater levels of mental health coverage and more provide coverage that is on par with coverage for other health concerns, i.e. mental health parity. These policies can help prevent suicide by increasing the accessibility and affordability, and ultimately the use of, needed mental health services. As more people access mental health services this helps normalize treatment seeking in the population, reduces of mental illnesses like depression and bipolar disorder, and in turn, reduce rates ofsuicide and suicide attempts. Potential Outcomes I Increased utilization of mental health services I Decreased of mental illnesses I Decreased rates of suicide attempts 0 Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. 0 Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the US. population that provides 15 data on substance use, mental health conditions, and services utilization. Using data from this survey, Harris, Carpenter, and Bao {2006) found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013} examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of 3.1 million per suicide prevented (Lang, 2013). 16 Establish Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment policy-level interventions, modifications to the environment) as well, increase the likelihood of success {Haddon, 1980}. Protective environments may be defined in part, as those where risk factors associated with suicide are limited and where protective factors are encouraged. Examples of risk factors include easy access to lethal means, stigma related to help-seeking, and substance abuse (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Protective factors include such things as social connectedness and access to mental health services (US. Office ofthe Surgeon General St National Action Alliance for Suicide Prevention, 2012). Establishing environments that address these factors where individuals live, work, and play, can help prevent suicide. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. 0 Reducing access to lethal means. Means of suicide such as firearms, hangingfsuffocation, or jumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1} the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes [Deisenhammer et al., 2009; Simon et al., 2001) and 2) that people tend notto substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 200?; Yip et al., 2012}. Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means: no intervening or Suicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may 17 include education and counseling around storing firearms-docked in a secure place in a gun safe or lock box), preferably unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 81 Rivara, 2016; C. W. Runyan et al., 2016). Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation). Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides {Escobedo 84 Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004). Potential Outcomes I Increase in safe storage of lethal means I Reduction in suicide attempts I Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016), as described below. Means restriction. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide {Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year [Perron, Burrows, Fournier, Perron, 8: Ouellet, 2013). Moreover, the reduction in suicides byjumping was 18 sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removai of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, sadly, both the number and rate of suicide increased fivefold {Beautrais, 2001; Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009). Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethai Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. in a pre-post quality improvement project, Runyan et. al (2016) found that at post-test 76% (of the 55% of parents follow-ed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 84 Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 73.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara Martin, 2012) Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the 19 culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service-wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8L Caine, 2003). Using a time-series design to examine the impact of the AFSPP program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post?launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010}. Community-based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8c Remer, 2009}. 20 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1951). Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others {Centers for Disease Control and Prevention, 2009}. Social connections can be formed within and between multiple levels of the social ecology (Dahlber E: Krug, 2002}, for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 81. Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013). Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, communityf neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Con nectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009) Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 21 Comment IAI: Linda, lstill might like to research this approach a bit more to find Community engagement activities. lCommunity engagement is an aspect of social capital. a more robust study. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean?up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes Reduction in maladaptive coping attitudes and behaviors Increase in healthy coping attitudes and behaviors Increase in referrals for youth in distressed Increase help?seeking behaviors Positive perception of adult support Evidence Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. :l Peer norm approaches. Evaluations show that programs such as Sources ofStrength can i improve school norms and beliefs about suicide that are created and disseminated by student i peers. In a randomized controlled trial oiSources ofStrength conducted with 18 high-schools i (E metropolitan, 12 rural], Wyman et al. (2010] found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a i history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in i maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010]. i Fommunity engagement activities}: vacant lot greening initiative was undertaken in Philadelphia between 1999 and LEiit'eIi?WW? together to green 4,436 lots [or 18 million square feet] in 4 areas of the city. Researchers found significant associated reductions in community residents? self~reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism {Branas et al., 2911). 22 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem- solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014). Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1936), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters {Pollock El. Williams, 2004). Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8L Bunney, 2002} and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, St Beck, 2012; Townsend et al., 2001) appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use) associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, 23 Comment IAI: Kristin, can you add what the acronym stands for and a sentence or two about the program? Comment IAI: Same comment as above.) F-il-h? including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse {le S. Knox, Burkhart, at Hunter, 2010). #35; Comment Same comment. Potential Outcomes I Reduction in suicide attempts and sulcide ideation 5 Enhanced knowledge of risk and protective factors associated with suicide ii I Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) I Improve and normalize help-seeking behavior Enhance social competence and emotional regulation skills i i Enhance problem-solving and conflict management skills Evidence There are a several programs with evidence that supports teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. ?1?h 0 Social emotional learning programs. In a cluster?randomized controlled trial of??AM ]conducted across 10 European Union countries and 163 schools, students participating in the YAM program were signi?cantly less likely to have an incident suicide attempt 0.45, 0.24~0.85; p=0.014l and severe suicidal ideation {0.50, 0.27-0.92; p=0.025) at the 12- ?i month follow?up compared to the control group. Additionally, related to severe suicide 3 1" ideation, in the TAM group absolute risk fell by 0.50% and RR fell by 49.5% {Wasserman et 3 i i i aL,2014t Additionally, in a randomized controlled trial, was shown to reduce self?reported suicide attempts at 3-months post intervention among participating students compared to i control students. The 505 program also increased students' knowledge of how to get help i for themselves or friends for depression andfor suicidal thoughts, and favorable attitudes i 5 toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling, Aseltine, St James, 2016]. rted half Finally, in an outcome evaluation of the the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 606 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of 686 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions 24 {C?mmenl? IAI: Sentence ortwo about .l the program is needed here. {Wilcox et al., 2003}. This finding likely arose due to the lack of implementation fidelity and i: {Comment IAI: Same comment. pointed to the need for 686 to be delivered with precision, consistency, and teacher support. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide [Kellam et al., 2008). :l 5 Parenting skill and family relationship programs. Parenting and family skills .l 3 i' I approaches have shown promising impacts in preventing key risk factors associated with suicide. Several studies have demonstrated the effect offrhe incredible Yearsbrogram on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster?Stratton, Reid, Si. Beauchaine, 21311; Webster-Stratton, Jamila Reid, Stoolmiller, 2003]. The program is also associated with improved problem-solving and conflict management; these skills were maintained at layear follow?up (Reid, Webster? Stratton, Hammond, 2003; C. Webster-Stratton Hammond, 199?; C. Webster-Stratton, Reid, 81 Hammond, Mill). The program demonstrated greater benefits as the dosage ofthe intervention increased [Herman et al., 2011]. alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families [Spoth, Guyll, 8: Day, 2002]. 25 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at- risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. Pepple who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults {and youth}. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t ?slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. 0 Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers and/or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of 26 depression, hopelessness, and subsequent mental health care utilization. Like means reduction, favor and create an endnote reference for crisis interventions can put space or time between an individual who mav be considering this in the endnote library called (Suicideenl) and then insert the ref here? Comment IAI: Kristin can you do me a I Thank vou! Comment Please add a sentence dEScribing MHFA 1 '{Consultant Please add sentence here describing the program. suicide and harmful behavior. r?h?s Potential Outcomes Reduction in suicide attempts I 0 Reduction in suicide deaths 0 Increased identification of individuals at?risk for suicidal behavior i i . Increased at-risk individuals in treatment Increased community members trained to Identify at?risk Individuals I: - Increased referrals for health care ii i f: 5 Evidence There is evidence that communitv gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be Abuse and Mental Health Services Administration, Center for Mental Health Services. Report to Congress: Garret Lee Smith-Suicide i Prevention Program. US Department of Heaith and Human Services: MD, 2014). However, .l there is limited evidence for effectiveness screening programs, but at the same time, standard 5' principles for public health screening make them promising {Pena Caine, 2006). The number of :l studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. It Gatekeeper training. In a randomized controlled trial of 300 participants ofEi/lental Heoith First }?Iid (MHFA), the intervention group reported greater confidence in providing help to others, 'gE'at'? 'r'liic'?i iE'g' in" Ffa' "it? ith i health professionals about treatments, and decreased stigmatizing attitudes. Additionallv, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. [Kitchener 3.: Jorm, 2004}. Additional research rigorouslv :l evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is i I needed [Kitchener 8t Jorm, 2006). I Screening combined with care management and overall continuity of care. An examination of the impact of the Penn} ford Perfect Depression Core (Fire-cursor to Zero Suicide} program . found that there was a dramatic and statisticallv significant decrease in the rate of suicide between the baseline vears prior to the intervention {1999 and 2000} to the intervention vears (2002?2009}. During this time period, the suicide rate fell 82% (C. E. Coffev, 2006; C. E. Coffev, 2? {Comment Hi: Maybe add a detail here Coffey, 8t Ahmedani, 2013]. Further, suicide rates also declined among HMO members who about ASIST participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 31 Ahmedani, 2015]. Crisis intervention. In an evaluation of the effectiveness of the National Suicide Prevention Lifeline to prevent suicide, 1,035 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow?up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that participants experienced significant decreases in suicidality over the course of the telephone session. and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, Kleinman, 200?). another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, lviunfakh, and Kleinman [2013) assessed the impact of the Applied Suicide intervention Skills Training a widely implemented gatekeeper training brogra across the NSPL network of hotlines over the period 2008?2009. Using data from 1,410 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were signi?cantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end oftheir call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 23 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co?worker, or other acquaintance to suicide lPitman, Osborn, King, 3: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer 8L Sonneck, 1993; Niederkrotenthaler 81 Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post?discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. 0 Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities le.g., home visits, mail, telephone, e-mail} to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296-2594: reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) 0 Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas Kutcher, 2011}. 29 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide and/or suicide contagion. Potential Outcomes Reduction in mental health?related sequelae Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suiciderrelated mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow?up, much of the evidence in this area primarilv focuses on risk and protective factors. - Treatment for people at-risk of suicide. significantlv improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24?months of follow-up lHunkeler et al., 2006; Unutzer et at, relative to patients who received care as usual. Collaborative Assessment and Management of Suicidalitv has been tested and supported in correlational studies {Jobes, 2012), in a varietv of inpatient and outpatient Improving Mood Promoting Access to Collaborative [l'reatmeniillMPACT] has been shown to {Comment IAI: Need sentences for the rest of the programs. Ugh. Sorrvsettings and in one REIT with several additional RCTs under wav. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. (Comtois et al., 2011). Dialectical Behavioral Therapy. in a randomized controlled trial of women with recent suicidal or self-injurious behavior, those receiving DBT were half as likely to make a suicide attempt at 30 two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined (Linehan et al., 2006}. Attachment-Based Family Therapy (ABFT). A randomized controlled trial of ABFT (Diamond et al., 2010) found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care (-4.37 vs. -2.34; .001; Additionally, a higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 46.2%; .006; (Diamond et al., 2010). Treatment to prevent re-attempts. Emergency Department Brief Intervention with Follow-up Visits. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from suicide relative to a treatment?as-usual group versus respectively; chi2 13.83, 0.001] (Fleischmann et al., 2008). Active follow-up contact approaches intended to prevent reattempts among patients that have been hospitalized and subsequently discharged for suicide attempts have been found in a meta- analysis conducted by Inagaki et al. (2015)to reduce reattempts by approximately 17% for up to 12 months post?discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of postcrisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Postvention. in a study by Visser, Comans, and Scuffham {2014], StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in which outreach to 31 suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) (J. Cerel 8: Campbell, 2008). Safe messaging following a suicide. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually (95% confidence interval: -149 to -13; -2.32, df 54, 0.024) in the Viennese subwav system (Niederkrotenthaler 3; Sonneck, 2007) 32 Sector Involvement Public health can piay an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business/la bor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016). The National Electronic Injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.), age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose {Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts It is also important at all levels ilocal, state, and federal} to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk 34 and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly,r over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community?level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion 35 References {See last 7 pages after Appendix?EndNete put them there because it thought the Appendixftable was part of the regular text; Didn?t want to cut and paste table for fear it?d get tie-formatted) 36 Appendix A: Summary of Strategies and Approaches to Prevent uicid Strategy litmustroacI-uf Program, Practice or Policyr Strengthen ?nancial security Best Available Evidence Suicide Suicide Attempts or ldeation Other Risk;f Protective Factors for Suicide Lead Sectors1 1 Comment I?ll help you complete the lead sector column. For the other columns, you just need to insert a check? mark based on the evidence vou describe In the narrative for a particular program or policy. For example, if the evidence shows impact on suicide, then put a check-mark in that column. If the studllr also found effects on risk or protective factors, then put a check-mark in that Strengthen Unemployment benefit programs Government economic (local, state, supports I . - I I FEderal) Housmg stabilizatlon polIcres The National Neighborhood Stabilization Government Pl" am (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parity Lows Healthca re care Government (local, state, Federal) Means restriction lntervening at hot spots Government (local, state, Establish t' :ederall protective ofe storage pro: toss (a tempts} .3 storage overnment . of firearms and (local, state, enwronments . . medication} Federal} Organizational policies and culture Lcolumn as well. Best Availabl Evidence Togetherfor Life Business/Labor US Air Force Suicide Prevention Program (family Government violence) (local, state, Federal) Business/Labor Community?based policies to reduce excessive alcohol use Aicohoi outiet density Government (local, state, Federal) Peer norm approaches Promote connectedness SDWCES of Strength Public Health to protect Social Services against suicide actwatles Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Manta} Heaith Program Public Health Social Services Signs of Suicide Teach co in 3 Good Behavior Game and problem- solving skills Parenting skill and family relationship approaches The incredibie Years Public Health Social Services Strengthening 10-14 Best Availabl Evidence Gatekeeper training Mentai Heaith First Aid Public Health Healthcare Social Services Identify and Screening combined with care management people Henry Ford Perfect Depression Care {Pre- Healthcare cursor to Zero Suicide) Crisis Intervention Nationai Suicide Prevention Lifeiine Public Health Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide improving Mood - Promoting Access to Healthcare Coiiaborative Treatment Social Services Intervene to Ufogcigg?gtrigjggessment and Management lessen harms and We?? Diaiecticai Behaviorai Therapy future risk Attachment-Based Famiiy Therapy Treatment to prevent rerattempts Best Availabl Evidence ED Brief intervention with Follow-up Visits Health ca re Active follow-up con tact approaches EST for Suicide Prevention Postvention Stono?By Response Service Healthcare Safe messaging following a suicide Medio Guidelines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as nongovernmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. REFERENCES Arsenault-Lapierre, (3., Kim, C., St Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Ba ndura, A. (1986). Socioifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., Simon, T. R. [2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), 336-347. Basile, K. C., Hamburger, M. E., Swahn, M. H., 8; Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14(4), 329. Beautrais, A. L. {2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust i'siZJt 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., 8: Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1127i1022801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. [2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i6?i11. Borowsky, I. Hogan, M., 84 Ireland, M. {1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A, MacDonald, J. M., Tam, v. W., Jackson, T. D., 8; Ten Have, T. R. (2011). A difference? in?differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 174(11), 1296?1306. doi:10.1093iajefkwr273 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M., Knoble, N. B., Shortt, J. W., 84 Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231-280. Carter, M., McGee, R., Taylor, 3., 3: Williams, S. Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51-62. Cava nagh, J. T., Carson, A. J., Sharpe, M., Lawrie, S. M. (2003). autopsy studies of suicide: a systematic review. Med, 33(3), 3954405. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full version-a.pdf Centers for Disease Control and Prevention. (2015a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2015e). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel, 1., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of "Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lhfe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. G., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-285. Classen, T. 1., 84 Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heuith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524-1526. Coffey, C. E., Coffey, M. 1., Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. ppi.PS.640422 Coffey, M, Coffey, C., 84 Ahmedani, B. K. (2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963372. doi:10.1002lda20895 Coulton, C. 1., Crampton, D. 5., Irwin, M, Spilsbury, J. C., 8: Korbin, J. E. How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negl', 111741142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. Crosby, A. E., Han, I3., Ortega, L., Parks, 5. E., 84 Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 55-13), 1-22. Crosby, A. E., Ortega, L., 8.: Melanson, C. (2011}. Self?directed Iv?iciience Uniform Definitions and Recommended Data Elements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 81 Sacks, 1. 1. (2002}. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav, 32(3), 321-328. Curtin, S. C., Warner, M, Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, 1., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am Epidemioi, 180(1), 45-52. 106 Dahlber, L. L., Krug, E. G. (2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, 1. A. Mercy, A. B. Zwi, 8: R. Lozano Worid Report on Violence and Heaith (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, 5., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., St Rosenheck, R. A. (2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am 162(2), 311-318. Diamond, G. 5., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., 5he ef, K., 8: Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1951). Suicide: A study in sociology (JA Spaulding G. Simpson, trans). Glencoe, ll.: Free Press.{Original work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 84. Vella-Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 3? countries. Adolesc Health, 45(4), 351-359. Escobedo, L. 6., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anai Prev, 34(5), 689694. Etzersdorfer, E., Sonneck, G. (1998). Preventing Suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. 0., Hindman J., Reed, G. A., 8: Iskander, J. {In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 203?709. doi:50042? 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 81 Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311-316. Freisthler, E., Merritt, D. H., El. LaScala, E. A. (2006). Understanding the ecology ofchild maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263?280. doi:10.1127l1072559506289524 Frieden, T. R. (2010). A framework for public health action: the health impact pyramid. Am Public Health, 100(4), 590-595. doi:10.21OSiAJ PH.2009.135552 Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. Ghahramanlou-Holloway, M., Bhar, 5., Brown, 6., Olsen, (3., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. 5., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M., 8: Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. H., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(5), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. doi:10.1521fsuli2001373338 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-214. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Harris, K. M., Carpenter, C., 8: Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8L Carter, 6. L. [2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Journai of 198(4), 309-316. Hawton, K. (2007}. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Heise, L., 8: Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, R. Lozano {Eds}, Worid Report on Vioience and Heaith (pp. 87-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 3: Webster?Stratton, C. (2011}. The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's (Io?Occurring Internalizing Q, 26(3), 189?201. doi:10.1037la0025228 Hong, J. 5., Kral, M. J., Espelage, D. L., St Allen?Mea res, P. {2012). The social ecology of adolescent?initiated parent abuse: 3 review of the literature. Chiid Hum Dev, 43(3), 431?454. 011?02234; Houle, J. N., 81 Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073-1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259?263. Inagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta~ana ysis. JAffect Disord, 125, 66-73. doi:10.1016fj.jad2014.12.048 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence; based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), BAD-653. doi:10.1111fj.1943? 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 81'. Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124-2133. Kellam, S. (3., Brown, C. H., Poduska, J. M., lalongo, N. S., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohoi Depend, 95 Suppi 1, 55-528. Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust 40(1), 6-8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 81'. Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. 5910la000136 Klevens, J., Simon, T. R., Chen, J. (2012}. Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 2.7110), 1987-2002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ, 1376. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457-2463. Knox, M. 5., Burkha rt, K., Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), ?3-88. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), Losel, F., 8 Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58?523. Luo, F., Florence, C. 5., Quispe?Agnoli, M, Ouyang, L., 81 Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 19282007. Am Public Health, 101(6), 1139?1146. doi:10.2105fajph2010300010 Maimon, D., Browning, C. R., Brooks~Gunn, J. (2610). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 301324. doi:10.117710022146510372878 Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., 81 Martin, N. {2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162?168. Mitchell, A. M, Kim, Y., Prigerson, H. (3., 8L Mortimer?Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12?18. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828?833. Muennig, P., Cohen, A. K., Palmer, A., El. Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18n26. National Action Alliance for Suicide Prevention. (2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention. (2016). Action Alliance Priorities. Retrieved from National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief; Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust ZJ 41. Olfson, M., Gerhard, T., Huang, C., 5., 81. Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 22(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journai of 181(3), 193-199. Pena, J. B., 8: Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614-637. doi:10.1521fsuli.2006.36.6.614 Perron, 5., Burrows, 5., Fournier, M., Perron, P. A., 8: Ouellet, F. [2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am} Puhiic Health, 103(2), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. (2012). The impact of neighborhoods on intimate partner violence and victimization. Trauma Viaierice Abuse, 13(2), 112-132. Pirkis, J., Too, L. 5., Spittal, M. J., Krysinska, K., Robinson, J., 84 Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, 0., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. H., 8: Williams, J. M. {2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, 0., McKee, M., Gunnell, 0., Chang, 5.5., St Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 18134814. Reid, M. J., Webster?Stratton, C., 8: Hammond, M. (2003). Follow?up of children who received the Incredible Years intervention for oppositional?defiant disorder: Maintenance and prediction of 2?year outcome. Behavior Therapy, 34(4), 411491. Resnick, M. 0., Ireland, M., 8: Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. Adoiesc Heaith, 35(5), 424 ed21? 410. Rowhani?Ra hba r, A., Simonetti, J. A., 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111424. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., ?t Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West.)f Emerg Med, 17(1), 8?14. Runyan, D., Wattam, C., Ikeda, R., Hassan, F., Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Loza no (Eds), Worid Report on Vioience and Heaith (pp. 5986). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., St Gannon?Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annuai Review ofSocioiogy, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 8: James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Shepard, D. 5., Gurewich, D., Lwin, A. K., Reed, (3. A., Jr., St Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. doi:10.1111isltb.12225 Silverman, M. M., 81. Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and Life- threatening behavior, 25(1), 10-21. Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8i O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., Day, 5. X. (2002). Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. 1 Stud Aicohoi, 63(2), 219- 228. Sudak, H., Maxim, K., 3; Carpenter, M. (2003). Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136-142. doi:10.1176iappi.ap.32.2.136 Szumilas, M., 3: Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Public Health, 102(1), 13-29. Townsend, E., Hawton, K., Altman, D. (3., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979-933. US. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. JArn Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journalof Community 42(1), 19-23. doi:10.1002ljcop.21536 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide the Threat Behav, 46(1), 106?120. Wasserman, D., Haven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, 0., . . . Carli, V. {2014). School? based suicide prevention programmes: The SEYLE cluster?randomised, controlled trial. The Lancet, sasieszz), 1536?1544. Webster?Stratton, (2., Hammond, M. (1997). Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consult Clin 65(1), 93?109. Webster?Stratton, (2., Reid, M. J., 34'. Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 233302. Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191*203. Webster?Stratton, (2., lamila Reid, M., 31 Stoolmiller, M. (2003). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schools. Journal of child and 49(5), 471?433. Widome, R., Sieving, R. E., Harpin, S. A., Hearst, M. O. (2003). Measuring neighborhood connection and the association with violence in young adolescents. lAdolesc Health, 43(5), 432-439. Wilcox, H. C., Kellam, S. (3., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., Anthony, J. C. (2003). The impact of two universal randomized first- and secondegrade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 56003. Wilson, W. l. (2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A., Brown, C. H., LoMurray, M., SchmeeIk-Cone, K., Petrova, M., Tu, . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Aml Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 31 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. A Technical Package to Prevent Suicide Prepared b? . Division ofViolence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 Comment Need to come up with a title for the package and insert it where Lindicated. '1 Com-em IAI: insert the names ofthe TP group members who contributed to i the development of the package assume evewone has contributed in some way. You and Kristin should be listed ?rst given all of your work on the package to Ldate. [Title] is a publication ofthe National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technicai Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. B20 ntentig Comment insert the table of contents; see other packages as example of how to structure this section External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness to protect against suicide; teaching caping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Suicide Prevention Strategy and the National Action Alliance for Suicide Prevention's priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8i Melanson, 2011). Self-directed violence may be suicidal or non-suicidoi in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotoi self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. in 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the US (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8: Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 13 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicide ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8r Krug, 2002): I Relationship Some risk Factors for suicide include (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014): Ea Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous Need to drop thefigure and write the factors up in narrative form. I suicide attempt, violence victimization, genetic and biological determinants, hopelessness I 0 Relationship: High conflict or violent relationships, sense of isolation and lack of sociai support, family/ioved one*s history of suicide, financial and work stress us Community: inadequate community connectedness. barriers to health care lack of access 1? I to providers and medications] Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma ,1 I I 1 associated with help-seeking and mental illness} It is important to recognize that the vast majority of individuals who are depressed (or who have other risk factors} do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio?cultural and economic status (0.5. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 201a] Protective factors, or these influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide. strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office ofthe Surgeon General 84 National Action Alliance for Suicide Prevention, 2012; World Health Organization, 201A). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 1 v'illaveces, 2004; Klevens, Simon, Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, 8: Rosenheck, 2005] as well as perpetration of child maltreatment [Coultom Crampton, lrwin, Spilsbury, 8L Korbin, 2007; Freisthler, Merritt, 8i. LaScaIa, 2005], teen dating violence (Capaldi, Knoble, Shortt, S: Kim, 2012}, intimate partner violence (Pinchevsky El. Wright, 2012), and youth violence (Sampson, Morenoff, 8i Gannon-Rowley, 2002}. Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Ouispe?Agnoli, Ouyang. 8t Crosby, 2011,- Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, Ikeda, Hassan, Ramiro, 2002}, intimate partner violence (Heise Garcia?Moreno, 2002, Pinchevsky 8f. Wright, 2012}, sexual violence (Centers for Disease Control and Prevention, 2016:} and youth violence (Wilson, 2011}. Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 20153, 2016c, ZlJl?e; US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012]. Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8i Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, 84 Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen-Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, Vella-Zarb, 2009; Maimon, Browning, Brooks-Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel 81 Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne- Maxim, 198?; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the US. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8L Sacks, 2002). Research indicates that the health consequences of violence, including suicide, are also much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt (Chapman Br Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8.: Mortimer-Stephens, 2004), stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, 8: Marshall, 2014; Sudak, Maxim, Carpenter, 2008). The economic toll of suicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (US. Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research 9 suggests that suicide will not be prevented by any single intervention taking place in any single setting {Silverman St Maris, 1995; U.S. Office of the Surgeon General 31 National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family-, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor) (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014). Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta?analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; cl meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness}. In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their 10 implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt]. Preventing Suicide Strategy Approach Strengthen economic supports - Strengthen household financial security Housing stabilization policies Strengthen access to mental health care I Coverage of mental health conditions in health insurance policies Create protective environments I Reducing access to lethal means among persons at-risk of suicide Organizational policies and culture Ir Community-based policies to reduce excessive alcohol use Promote connectedness to protect against Peer norm approaches suicide Ir Community engagement activities Teach coping and problem-solving skills Social-emotional learning Parenting skill and family relationship approaches Gatekeeper training Screening combined with care management Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re~attempts Postvention I Safe messaging following a suicide Identify and support people at risk Intervene to lessen harms and prevent future risk I'll. The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and 11 of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Suicide ideation, attempts, morbidity and mortality vary by gender, racefethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as businessllabor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale . [Public health research suggests that some ofthe biggest impacts on health come from policies developed to improve socioeconomic conditions {Frieden, 2010}. Downturns in the economy and increases in unemployment and home foreclosures are associated with increased rates of suicide {Fowlen Gladden, ll Vagi, Barnes, 8i Frazier, 2015; Luo et al., 2011]. Policies that strengthen economic supports can help people stay in their homes, pay for necessities such as food and medical care, and getjob training, among it other things. In providing this support, stress and anxiety and the potential for a crisis situation may be Approaches [Economic and housing supports for individuals and families can be strengthened by improving policies that enhance financial security and stabilize housing for people, especially in times of economic need] I Strengthen household financial security. Etudies from the U5. indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 54 years old {Luo et al., 2011; 1. I. Fowler et al., 2015).] Potential {Jutcolnes - Reduced suicide rates 0 Lower foreclosure rates a Lower eviction rates 0 Reduced emotional distress Evidence There is evidence suggesting that strengtheni_ng household financial security and stabilizi_ng housing can reduce suicide risk. Strengthen household ?nancial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits lelus, Glymour, Avendano, 2014]. Another U.S. study examining the link between unemployment Comment IAI: The rationale should focus on the strategy - strengthening economic supports - and how and why that might be beneficial in reducing suicide. As part ofthe rationale you can talk about how suicide relates to economic factors leg, historical data shows that suicide rates rise during periods of economic recessions and decreases during periods of economic expansion]: how economic stressors {job loss, long periods of unemployment, reduced income, difficulty covering medical, food, housing expenses, etc.l increase risk for suicide and how buffering these risks can potentially JJ?rotect against suicide. You cover this a Comment Statement should be more along the lines of: ?There are a number of approaches that can help strengthen economic supports and buffer against the risk for suicide." _economic stress." Or ?Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of J5. Comment This would be a good introductory sentence to the rationale. For the two approaches, suggest saying something along the following lines: Strengthening household ?nancial security can reduce the risk for suicide by providing individuals with the necessary means to cover basic expenses and lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits, livable wages, Temporary Assistance to Needy Families medical bene?ts, and retirement and disability insurance to help cover the costs of basic necessities or to offset costs 1 and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed tojust the loss of job, predicted suicide risk {Classen 3i Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to 13 strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance) have also shown an impact on suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied. At the national level, an estimated 3,000 fewer suicides would occur per year nationwide if every state increased their per capita spending on these types of assistance by $45 per year (Flavin 8: Radcliff, 2009]. Housing stabilization policies. The Nationall Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults {Houle Light, 2014). Another study of data from 16 US. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began} to 2010 (after it had peaked; Fowler et al. 2015). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 14 Strengthen Access to Mental Health Care Rationale While most people with mental health problems do not attempt or die bv suicide (Olfson, Gerhard, Huang, 8i. Stroup, 2015], previous research indicates that mental illness is an important risk factor for suicide (WHO, 2am}. Studies suggest that up to 90% of people who die bv suicide may have had a . mental illness at the time of their deaths ECavanagh, Carson, Sharpe, Lawrie, 2003i State-level suicide rates have also been found to be correlated with general mental health measures such as depression {ArsenauIt-Lapierre, Kim, 8t Turecki, 2004]. Findings from the National Comorbiditv Survev indicate that relatively' few people in the U.S. with mental health disorders receive treatment for those conditions {Kessler et al., [200 . Lack of access to mental health care is one of the contributing factors related to the underuse of timely, affordable, and qualitv mental health care for people in need is a critical component to suicide prevention (WHO, 2014]. Apart from the treatment benefits, it can also serve to normalize help?seeking i behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns li.e., mental health paritv}. Benefits and services covered include such things as the number of visits, co-pavs, deductibles, inpatient/outpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health paritv law than the federal pariti,r law, then insurance plans regulated by the state must follow the state paritv law. Federal paritv replaces the state law only in cases where the state law prevents the application of the federal paritv law le.g., includes coverage for some mental health I ?qL? 1? a 1 Comment IAI: Add more citations here. You could also add a sentence indicating that suicide risk varies by type of disorder and the presence of other co?morbidities such as alcohol use disorders - see pg 40 Lin the World Suicide Report 2014 Comment IAI: Kessler RC, Demler Frank FIG, Difson lvl, Pincus HA, Walters EE, Wang P, Wells KB, Zaslavskv AM. Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med. Jun Comment Cunningham, P.J. Bevond Parity: Primarv Care thsicians' Perspectives On Access To Mental Health Care. Health Affairs, 28(3), weed-WWI. doi; originallv Jo. .2 conditions but not others]. Equal coverage does not necessarily:r implv good coverage as health insurance plans varv in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes Increased utilization of mental health services Decreased of mental illnesses Decreased rates of suicide attempts Decreased rates of suicide 15 published online April 14, 2009 Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the US. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, Harris, Carpenter, and Bao {2006} found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of SIB-3.1 million per suicide prevented (Lang, 2013). 16 Create Protective Environments Elationals] Prevention efforts that focus not only on individual behavior change log, help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes lHaddon, 1980}. Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide. Such approaches can involve changes to organizational climates, policy changes, and changes to the physical and social environment to reduce risk characteristics, encourage help-seeking and increase other protective factors. Modifying characteristics of the environment helps to create a context that promotes positive behavior and limits harmful behavior. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. - Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hangingfsuffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that lithe interval between thinking about and attempting suicide can be as short as 5 or minutes (Deisenhammer et al., 2009; Simon et al., 2001) and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, Yip et al., 2012]. Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help (Cox et al., 2013]. 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable andfor impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms??locked in a secure place in a gun safe or lock box], unloaded and separate from the ammunition--and keeping medicines in a 1? ?x Comment IAI: I edited the text in the rationale to more closely align with the notion of creating a protective environment. More could probably be added in terms of the research, but this gives you a sense of what the rationale for this section might look like. locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts (Rowhani?Rahbar, Simonetti, St Rivara, 2016; C. W. Runyan et al., 2016}. Organizational policies and culture that promote protective environments may be implemented in places ofemployment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation]. Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8; Ortiz, 2002; Giesbrecht et al., 2015]. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, Wilcox, 2004). Potential Outcomes I Increase in safe storage of means 0 Reduction in suicide attempts I Reduction in suicide deaths 0 Increase in help-seeking 0 Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016}, as described below. Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide {Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year {Perron, Burrows, Fournier, Perron, 8a. Duellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 18 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removai of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal ofthe barrier, sadly, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009]. Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. {2005} found that storing firearms unloaded, separate from ammunition, in a locked place andfor secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethai Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et. al {2016} found that at post- test 76% (of the 55% of parents followed up, n=114l reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post?test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara St Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000) (Mishara 8: Martin, 20121 Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents 19 of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 81. Caine, 2003}. Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010]. These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010}. Community-based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol?involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, Remer, 2009). 20 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002}, for instance between individuals peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one's community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community,?l neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all ofwhich helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 21 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes - Reduction in maladaptive coping attitudes and behaviors 0 Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed I Increase help-seeking behaviors I Positive perception ofadult support Eviden ce Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (6 metropolitan, 12 rural], Wyman et al. (2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism [Branas et al., 2011). 22 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1985), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters (Pollock Williams, 2004}. Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8; Bunney, 2002) and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, 8: Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 23 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 81 Hunter, 2010]. Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program {Wasserman et al., 2014). In a cluster- randomized controlled trial of YAM conducted across 10 European Union countries and 163 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=D.025] at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% (Wasserman et al., 2014). Signs ofSuicide {505) is another school-based prevention program for students aged 13:11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 8: James, 2016}. In a randomized controlled trial, 505 was shown to reduce self-reported suicide attempts at 3-months post intervention among 24 participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression andlor suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling et al., 2016). Finally, the Good Behavior Game (636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the GBG is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2003). In an outcome evaluation of the 656, first graders assigned to GBG reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (386 effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between (336 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for (536 to be delivered with precision, consistency, and teacher support. 656 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years NY) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide) (Herman et al., 2011). The program includes 9- 20 sessions offered in communitynbased settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, lamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8; Hammond, 25 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionallv, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 26 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims ofviolence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults (and youth). I Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t 'slip through the cracks?. These approaches typically employ screening for depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotiine, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 27 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes 0 Reduction in suicide attempts - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at-risk individuals in treatment I Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained {Substance Abuse and Mental Health Services Administration, 2014]. However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. - Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid (MHFA) program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders}, possible riskfactors, and where and how to get evidence-based effective help (Kitchener Jorm, 2004}. In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener St Jorm, 2004]. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener 8i. Jorm, 2006). Gatekeeper training has also been a core part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the 28 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GL5 trainings had significantly lower youth suicide rates the year following the training implementation (-1.02, p=.03) (Walrath, et al., 2015) This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation of the GLS program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% CI, 1.8-8.0 fewer attempts per 1000 youths]; .003; Godoy Garraza et al., 2015). More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GL5 program. Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Care program was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up ca re system wide (C. E. Coffey, 2006}. An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 81 Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 8L Ahmedani, 2015). Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the National Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow?up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that participants experienced significant decreases in suicidality over the course of the 29 telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). in another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013] assessed the impact of the Applied Suicide intervention Training (ASST), a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in ASIST were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call to the hotline. Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 30 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide (Pitman, Osborn, King, 8: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer E: Sonneck, 1998; Niederkrotenthaler Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%_25% reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas 8: Kutcher, 2011}. 31 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andfor suicide contagion. Comment IAI: The Write?up of the evidence for this strategy needs to follow the format used in the other sections of the TP. I modified the write-up for the first approach to illustrate the form at. You will need to do something similar for the second approach. Potential Outcomes Reduction in mental health?related sequelae . Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow~up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of Impact on risk and protective factors for suicide. One example is the Improving Mood? Promoting Access to Coiiaborotive Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. it facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et ai., 2006}. The program has been shown to significantly improve quality.r of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2005; Unutzer et al., 2006?; relative to patients who received care as usual. Another example is Collaborative Assessment and Management ofSuicidaiity which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 32 patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (Jobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. {Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy (AFBT). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined (Linehan et al., 2006]. ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety [Diamond at al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care {-4.37 vs. -2.34; .001; Additionally, a higher percentage participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks {82.1% vs. 46.2%; .006; {Diamond et al., 2010). Treatment to prevent re-attempts. Emergency Department Brief Intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths 33 from suicide relative to a treatment-as-usual group versus respectively; chi2 13.83, 0.001} {Fleischmann et al., 2008}. Active follow-up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years}. These approaches have been found in a meta-analysis conducted by Inagaki et al. (2015} to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta?analysis were small, it was not possible to determine the effect of active contact and follow- up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001} found that patients who refused ongoing care but who were randomized to be contacted by letterfourtimes per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam, Sarjami, Kolahi, Carter, 2011; Wang et al., 2016}. Cognitive Behavior Therapy for Suicide Prevention uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and CST-5P also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial found of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005}. Postvention programs such as Stono?By Response Service are implemented with the goal of providing support to survivors of suicide. StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, 8; Scuffham, 2014}. In a study by Visser et al. (2014}, StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in 34 which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) (J. Cerel Campbell, 2008). Safe messaging following a suicide. Media guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact ofthe guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually [95% confidence interval: -149 to -2.32, df 54, 0.024) in the Viennese subway system (Niederkrotenthaler Sonneck, 2007) 35 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. 36 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System?All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 37 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, more timely surveillance data, and critical mention in the President?s FY17 budget, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking), and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination?-in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re-attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer 38 levels of the social ecologv, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 39 References ArsenauIt-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. doi: 10.1136f1471-244X-4-3? Bandura, A. (1986}. Sociaifounciations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), 336-347. Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journal of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi: 10.1080f0004867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M, Horwood, L. J., 8i. Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi: 10.1080;00048670902873714 Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi: 10.1177f1077801214554075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. doi: 10.1136Iinjuryprev-2015-041819 Borowsky, l. W., Hogan, M, Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), ET. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., 8: Ten Have, T. R. (2011). A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. doi: 10.1093/aje/kwr273 Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi: 10.1001ijama.294.5.563 Butchart, A., Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. doi: 10.1853111945656032.231 Carter, M, McGee, R., Taylor, 3., Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. doi: Cavanagh, J. T., Carson, A. J., Sharpe, M, 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. from Available at: Strategic Direction Full version-agndf Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016:). Sexual Violence: Risk and Protective Factors. from Available at: 40' Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2015e). Youth Violence: Risk and Protective Factors. from Available at: Cerel,1., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. doi: 10.1521isuli.2008.38.1.30 Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of "Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lhfe-threatening behavior, 44(6), 591-600. doi: Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Threat Behav, 37(5), 543-552. doi: 10.1521/suli.2007.37.5.543 Cherpitel, C. 1., Borges, G. L. G., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-285. Classen, T. 1., 84 Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heuith Econ, 21(3), 338-350. doi: 10.1002fhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524-1526. doi: 10.1176ips.2006.5?.10.1524 Coffey, C. E., Coffey, M. 1., 8i. Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. doi: 10.1176/appi.PS.640422 Coffey, M, Coffey, C., 84 Ahmedani, B. K. (2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. doi: Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963372. doi: 10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M, Spilsbury, J. C., 8: Korbin, J. E. How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negl', 111241142. doi: Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. doi: Crosby, A. E., Han, I3., Ortega, L., Parks, 5. E., 84 Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 55-13), 1-22. Crosby, A. E., Ortega, L., 8.: Melanson, C. (2011}. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Crosby, A. E., 8: Sacks, 1. 1. {2002). Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav, 32(3), 321-328. Curtin, S. C., Warner, M., 8: Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., 8-1. Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi: 10.1093/aje/kwu106 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, 1. A. Mercy, A. B. Zwi, at R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 1-56}. Geneva, Switzerland: World Health Organization. 41 DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, l. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. (2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. AmiI 162(2), 311-318. doi: 10.1176/appi.ajp.162.2.311 Diamond, G. 5., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 8: Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 3; G. Simpson, trans). Glencae, it: Free Press.{Original work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 84. Vella-Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 3? countries. lAdolesc Health, 45(4), 351-359. doi: Escobedo, L. 6., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 689694. Etzersdorfer, E., Sonneck, G. (1998). Preventing Suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. 0., Hindman 1., Reed, G. A., 8: Iskander, J. {In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 3: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195209. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi: 50042~ 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., Frazier, L. (2015}. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311-316. doi: 10.21051AJ PH.2014.301945 Freisthler, B., Merritt, D. H., 8i. LaScala, E. A. (2006). Understanding the ecology ofchild maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263280. doi: 10.117211077559506289524 Frieden, T. R. (2010). A framework for public health action: the health impact pyramid. Am Public Health, 100(4), 590-595. doi: 10.21051AJPH.2009.185652 Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am 1 Public Health, 104(1), 17-22. doi: Ghahramanlou-Holloway, M., Bhar, 5., Brown, 6., Olsen, (2., 8; Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. doi: 10.1111fadd.12762 Goldsmith, 5. K., Pellmar, T. C., Kleinman, A. M., 31. Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. 42 Gould, M. 5., Cross, W., Pisani, A. H., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(5), Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. doi: 10.1521fsu i.2007.37.3.333 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-214. doi: 10.1001fjama.293.6.707 Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Harris, K. M., Carpenter, C., 8: Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. doi: Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8L Carter, 6. L. {2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Journai of 198(4), 309-316. Hawton, K. {2007}. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi: Heise, L., 8: Garcia-Moreno, C. {2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, R. Lozano {Eds}, Worid Report on Vioience and Heaith (pp. 87-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M, 3: Webster?Stratton, C. (2011}. The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's (Io?Occurring Internalizing (2, 26(3), 189?201. doi: 10.1037fa0025228 Hong, J. 5., Kral, M. J., Espelage, D. L., St Allen?Mea res, P. {2012). The social ecology of adolescent?initiated parent abuse: 3 review of the literature. Chiid Hum Dev, 43(3), 431?454. doi: 10.1002/510528? 011?02234; Hunkeler, E. M, Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259?263. doi: 10.1136fbmj.38683.210255.BE Inagaki, M, Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a metamanalysis. JAffect Disord, 175, 66?73. doi: 10.1016fj.jad.2014.12.048 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence? based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi: 10.1111lj.1943- 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 81 Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124*2133. doi: Kellam, S. 6., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohoi Depend, 95 Suppi 1, 55-528. doi: Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. doi: 10.1186f1471-244X-4-23 Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust ZJ 40(1), 6-8. doi: Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. {2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi: 10.1022/0227- 59101a000136 43 Klevens, J., Simon, T. R., 81 Chen, J. (2012). Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. interpers Violence, 27(10), 1987-2002. doi: 10.1177i0886260511431441 Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 32212428), 1326. Knox, K. L., Pflanz, S., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457-2463. doi: 10.2105fajph.2009.159871 Knox, M. S., Burkha rt, K., St Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 23-88. doi: 10.1002lhec.1816 Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. doi: Losel, F., Farrington, D. P. [2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58?523. doi: 10.1016/j.amepre.2012.04.029 Luo, F., Florence, C. 5., QuispewAgnoli, M, Ouyang, L., 81 Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928?2002. Am Public Health, 101(6), 1139?1146. doi: 10.2105iajph.2010.300010 Maimon, D., Browning, C. R., 81 Brooks?Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307?324. doi: 10.11721'0022146510372878 Mishara, B. L. (1995). The impact ofSuicia?e. New York: Springer. Mishara, B. L., 8t Martin, N. (2012}. Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162468. Mitchell, A. M, Kim, Y., Prigerson, H. G., 8; Mortimer?Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, Motto, J. A., Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828?833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. {2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. doi: National Action Alliance for Suicide Prevention. (2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention. (2016). Action Alliance Priorities. Retrieved July 12, 2016, from National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust ZJ 41. doi: 10.1080i00048670701266680 Olfson, M., Gerhard, T., Huang, C., 5., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 22(12), 1172-1181. doi: Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193?199. 44 Pena, J. B., 81 Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614-632. doi: 10.1521f5u i.2006.36.6.514 Perron, S., Burrows, S., Fournier, M., Perron, P. A., 8: Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Heaith, 103(7), 1235-1239. doi: 10.2105!ajph.2012.301089 Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma 1Lr?ioi'ence Abuse, 13(2), 112-132. doi: 10.1172/15248380124456441 Pirkis, J., Too, L. 5., Spittal, M. J., Krysinska, K., Robinson, J., 8: Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. doi: Pitman, A., Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. R., 3: Williams, J. M. {2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, D., Chang, S. 5., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. doi: 2 Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 411?491. Resnick, M. 0., Ireland, M., 8: Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdoiesc Heaith, 35(5), 424 e421? Rowhani?Ra hba r, A., Simonetti, J. A., 8: Rivara, F. P. (2016). Effectiveness of interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111?124. doi: Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., 81 Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med, 17(1), 8?14. doi: 10.5811iwestjem.2015.11.28590 Runyan, D., Wattam, C., Ikeda, R., Hassan, F., 81 Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, at R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., Gannon?Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annuai Review ofSocioiogy, 28, 443428. Schilling, E. A., Aseltine, R. H., Jr., 8; James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 12(2), 152?166. doi: Shepard, D. 5., Gurewich, D., Lwin, A. K., Reed, G. A., Jr., 81 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. doi:10.1111isltb.12225 Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and iife- threatening behavior, 25(1), 10-21. Simon, 0. R., Swarm, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., Day, S. X. (2002). Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. Stud Aicohoi, 63(2), 219- 223. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., 8; Carpenter, M. {2008). Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136-142. doi: 10.1176/appi.ap.32.2.136 45 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Public Health, 102(1), 13-29. Townsend, E., Hawton, K., Altman, D. (3., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979-933. U.S. Office of the Surgeon General, 3: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. 1.4m Geriatr Soc, 54(10), 1550-1556. Visser, V. 5., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. lournalof Community 42(1), 19-23. doi: 10.1002/jcop.21536 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. doi: 10.1111/sltb.12177 Wasserman, D., Haven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster?randomised, controlled trial. The Lancet, 385(9977), 1536?1544. doi: Webster?Stratton, C., 81 Hammond, M. Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consult Clin 65(1), 93?109. Webster?Stratton, (2., Reid, M. J., Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 2333302. doi: Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child 3 Adolescent 40(2), 191003. Webster?Stratton, C., Jamila Reid, M., Stoolmiller, M. (2003). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in highsrisk schools. Journal of child and 49(5), 471?433. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2003). Measuring neighborhood connection and the association with violence in young adolescents. JAdolesc Health, 43(5), 432439. doi: Wilcox, H. C., Kellam, S. (3., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., 8: Anthony, J. C. (2003). The impact of two universal randomized first- and classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?73. Wilson, W. J. (2011). When warlr disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A., Brown, C. H., LoMurray, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8L Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. doi: 46 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Approach/Program, Practice or Policy Suicide Suicide Other Lead Sectors1 Attempts or Riskarotective ldeation Factors for Suicide Strengthen household financial security Strengthen Unemployment bene?t programs 1/ Government economic (local, state, supports Federal) Housing stabilization policies The National Neighborhood Stabilization 1/ Government Program (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parityr Laws v/ v? Government care (state, Federal) Health care Reducing access to lethal means among persons at?risk intervening at hot spots Government (local, state, Establish 5 :eiira: protective afe storage practices 1/ u: ea environments Organizational policies and culture Togetherfor Life 1? Businesleabor US Air Force Suicide Prevention Program Best Available Ev' - Government (local, state, Federal) Businesstabor Community?based policies to reduce excessive alcohol use against suicide Alcohol outiet density v? Government (local, state, Federal) Peer norm approaches Promote connectedness Sources 0f Strength Public Health to protect Social Services Community-engagement activities Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Mental Health Program Public Health Signs of Suicide Education Good Behavior Game solving skills Parenting skill and family relationship approaches The incredible Years Public Health Education Strengthening 10-14 Best Available Evidence identify and Gatekeeper training Mental Health First Aid Public Health Healthcare Social Services Screening combined with care management support people Henry Ford Perfect Depression Core (Pre- Health ca re at-risk cursor to Zero Suicide) Crisis Intervention Notionoi Suicide Prevention Lifeiine v? Public Health Social Services Appiieo' Suicide intervention Training Treatment for people at risk of suicide improving Mood Promoting Access to in? Healthcare Coiioborotive Treatment Social Services Coiioborotive Assessment and Management of Suicidoiity CA M3) -st Available Evidence Interue no to Dialectical Behaviorai Therapy 1/ v" lessen harms and prevent Attachment-Based Famiiy Therapy 1/ future risk Treatment to prevent re-attempts ED Brief interven tion with Follow-up Visits 1/ Health ca re Active folio w?up con tact approaches 1/ 1/ CBTfor Suicide Prevention Postvention 5 tandBy Response Service 1" Health ca re Safe messaging following a suicide Media Guideiines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. A Technical Package to Prevent Suicide Prepared by: Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 [Title] is a publication of the National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technical Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. Broadly, the strategies represented include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term effects of suicidal behavior for individuals, families, communities, and society. Specifically, the strategies include strengthening economic supports; strengthening access to mental health care; establishing protective environments; promoting connectedness to protect against suicide; teaching coping and problem-solving skills; identifying and supporting people at?risk; and intervening to lessen harms and prevent future risk. This package supports the National Strategy for Suicide prevention, Goal 1, "Integrate and coordinate suicide prevention activities across multiple sectors and settings." {p.29} It also supports the National Action Alliance for Suicide Prevention?s priority (2016) "To create and disseminate a framework for comprehensive community-based suicide prevention." Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome (Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision- making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury (Crosby, Ortega, Melanson, 2011}. Self- directed violence may be suicidal or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended. Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotoi self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 in the U.S., the most recent death data available, suicide was responsible for 42,723 deaths, which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the tenth leading cause of death and has been among the top twelve leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 {Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 {Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behaviors (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide (Le. ideation) (Ferdon et al., in press). Suicides, suicide attempts, and ideation take an immense emotional, physical, and economic toll (see p. 9) on individuals, families and communities. By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8: Sacks, 2002). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002): Relationship Individual Some risk Factors for suicide include Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014): 0 Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness 0 Relationship: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community: Inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness It is important to recognize that the vast majority of individuals who are depressed (or who have other risk factors) do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General El. National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014) Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General at National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. As indicated above, suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes {Butchart, Phinney, Check, Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide {Desai, Oausey, 8i. Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, Irwin, Spilsbury, 8: Korbin, 2007; Freisthler, Merritt, 8: LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8: Kim, 2012), intimate partner violence (Pinchevsky 8: Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, 8: Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, Ikeda, Hassan, 8: Ramiro, 2002), intimate partner violence (Heise 8: Garcia-Moreno, 2002; Pinchevsky 8: Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, 8: Weingarden, 2012; Pinchevsky 8: Wright, 2012; Widome, Sieving, Harpin, 8: Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8: Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, 8: Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen- Meares, 2012; Losel 8: Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8: VeIIa-Zarb, 2009; Maimon, Browning, 8: Brooks- Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8: Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel 8: Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne-Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). Research indicates that the health consequences of violence, including suicide, are much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long?term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, 8: Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2008). The economic toll of suicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8i. Silverman, 2016}. The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million {Shepard et al., 2016]. Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (US. Public Health Service, 1999}. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman 8L Maris, 1995; US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare,justice, education, labor) {National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. According Frieden {2014), successful public health programs also require political commitment, funding, communication, and performance monitoring Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous le.g., RCT or quasi-experimental design} evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility ofimplementation in a US. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness}. In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts} provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and 10 family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality oftheir implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below}. The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen financial security Housing stabilization policies Strengthen access to mental health care - Coverage of mental health conditions in health insurance policies Establish protective environments a Means restriction I Organizational policies and culture I Community-based policies to reduce excessive alcohol use Promote connectedness to protect against I Peer norm approaches suicide I Community engagement activities 11 Teach coping and problem-solving skills - Socialaemotional learning 0 Parenting skill and family relationship approaches Identify and support people at risk - Gatekeeper training Screening combined with care management Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re?attempts I Postvention - Safe messaging following a suicide Intervene to lessen harms and prevent future risk The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Suicide ideation, attempts, morbidity and mortality vary by gender, racex?ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business/labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Public health research suggests that some ofthe biggest impacts on health come from policies developed to improve socioeconomic conditions (Frieden, 2010). Downturns in the economy and increases in unemployment and home foreclosures are associated with increased rates of suicide (Fowler, Gladden, Vagi, Barnes, Frazier, 2015; Luo et al., 2011). Policies that strengthen financial security can help people stay in their homes, pay for necessities such as food and medical care, and get job training, among other things. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Approaches Economic and housing supports for individuals and families can be strengthened by improving policies that enhance financial security and stabilize housing for people, especially in times of economic need. II Strengthen household financial security. Studies from the U.S. indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015). Policies that support financial security, such as provision of unemployment benefits, family assistance, education and training assistance, and other social welfare payments may help mitigate the risk of economic crises on suicide rates. Potential Outcomes 0 Reduced suicide rates I Lower foreclosure rates II Lovver eviction rates I Reduced emotional distress Evidence There is evidence that policies that strengthen household financial security and that stabilize housing can reduce suicide risk. I Strengthen household financial security. Provision of Unemployment bene?ts are one means to strengthen financial security. These benefit programs provide income protection during periods of unemployment in an effort to prevent or lessen the economic hardship for those experiencingjob loss and enduring unemployment. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment 13 benefits (Cylus, Glymour, &Avendano, 2014). Another U.S. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk (Classen 8: Dunn, 2012]. Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other social spending policies and programs have also shown an impact on suicide. A study by Flavin and Radcliff {2009) examined the impact of states? per capita spending on transfer payments retirement and disability insurance benefits, income maintenance benefits, unemployment insurance compensation), medical benefits, family assistance food and welfare programs), and total state spending on suicide rates between 1990-2000. A variety of suicide risk factors at the state level were also controlled in the analysis. What they found was that as per capita spending on total transfer payments, medical benefits, and family assistance increase there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on social programs specifically. In terms of lives saved, it was estimated that 3,000 fewer suicides would occur per year nationwide (Flavin 8: Radcliff, 2009). More research is needed to assess the impact of these policies in today?s social and economic context, but evidence thus far is promising. 14 Strengthen Access to Mental Health Care Rationale Mental illness is a risk factor for suicide-- Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths (Cavanagh, Carson, Sharpe, El Lawrie, 2003} and research on state-level suicide rates have been found to be correlated with general mental health measures such as depression (Arsenault-Lapierre, Kim, Turecki, 2004}. While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, 8: Stroup, 2015) and most people who have attempted suicide will not go on to die by suicide (Owens, 2002}, assuring access to affordable mental health care for people in need is critical to suicide prevention. Approaches A major approach to strengthening access to mental health care is through the provision of health insurance policies that include coverage for such services. - Coverage of mental health conditions in health insurance policies. Historically, mental health care was viewed as separate from physical health care and many health insurance plans either did not provide coverage of mental health services or such services were available but not affordable-due to higher co-pays or co-insura nce?- or set limits on the number of visits allowed. More recently, improvements have been made with many health insurance policies providing greater levels of mental health coverage and more provide coverage that is on par with coverage for other health concerns, i.e. mental health parity. These policies can help prevent suicide by increasing the accessibility and affordability, and ultimately the use of, needed mental health services. As more people access mental health services this helps normalize treatment seeking in the population, reduces of mental illnesses like depression and bipolar disorder, and in turn, reduce rates ofsuicide and suicide attempts. Potential Outcomes I Increased utilization of mental health services I Decreased of mental illnesses I Decreased rates of suicide attempts 0 Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. 0 Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the US. population that provides 15 data on substance use, mental health conditions, and services utilization. Using data from this survey, Harris, Carpenter, and Bao {2006) found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013} examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of 3.1 million per suicide prevented (Lang, 2013). 16 Establish Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment policy-level interventions, modifications to the environment) as well, increase the likelihood of success {Haddon, 1980}. Protective environments may be defined in part, as those where risk factors associated with suicide are limited and where protective factors are encouraged. Examples of risk factors include easy access to lethal means, stigma related to help-seeking, and substance abuse (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Protective factors include such things as social connectedness and access to mental health services (US. Office ofthe Surgeon General St National Action Alliance for Suicide Prevention, 2012). Establishing environments that address these factors where individuals live, work, and play, can help prevent suicide. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. 0 Reducing access to lethal means. Means of suicide such as firearms, hangingfsuffocation, or jumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1} the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes [Deisenhammer et al., 2009; Simon et al., 2001) and 2) that people tend notto substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 200?; Yip et al., 2012}. Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means: no intervening or Suicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may 17 include education and counseling around storing firearms-docked in a secure place in a gun safe or lock box), preferably unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 81 Rivara, 2016; C. W. Runyan et al., 2016). Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation). Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides {Escobedo 84 Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004). Potential Outcomes I Increase in safe storage of lethal means I Reduction in suicide attempts I Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016), as described below. Means restriction. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide {Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year [Perron, Burrows, Fournier, Perron, 8: Ouellet, 2013). Moreover, the reduction in suicides byjumping was 18 sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removai of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, sadly, both the number and rate of suicide increased fivefold {Beautrais, 2001; Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009). Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethai Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. in a pre-post quality improvement project, Runyan et. al (2016) found that at post-test 76% (of the 55% of parents follow-ed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 84 Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 73.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara Martin, 2012) Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the 19 culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service-wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8L Caine, 2003). Using a time-series design to examine the impact of the AFSPP program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post?launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010}. Community-based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8c Remer, 2009}. 20 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951). Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others {Centers for Disease Control and Prevention, 2009}. Social connections can be formed within and between multiple levels of the social ecology (Dahlberg 8: Krug, 2002}, for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 81. Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013). Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, communityf neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Con nectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009) Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 21 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes Reduction in maladaptive coping attitudes and behaviors Increase in healthy coping attitudes and behaviors Increase in referrals for youth in distressed Increase help-seeking behaviors Positive perception ofadult support Evidence Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (6 metropolitan, 12 rural}, Wyman et al. {2010) found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010). Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet} in 4 areas of the city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism (Branas et al., 2011}. 22 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem- solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014). Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1936), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters {Pollock El. Williams, 2004). Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8L Bunney, 2002} and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, St Beck, 2012; Townsend et al., 2001) appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use) associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, 23 including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse S. Knox, Burkhart, 8: Hunter, 2010). Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware ofMentoi Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program (Wasserman et al., 2014). in a cluster-randomized controlled trial of YAM conducted across 10 European Union countries and 168 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 95%Cl 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=0.025} at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% {Wasserman et al., 2014). Signs ofSuicio'e is another school?based prevention program for students aged 13-11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help-seeking behavior (Schilling, Aseltine, 3: James, 2016). In a randomized controlled trial, $05 was shown to reduce self-reported suicide attempts at 3-months post intervention 24 among participating students compared to control students. The 505 program also increased students? knowledge of how to get help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help-seeking. SOS participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower-risk participants (Schilling et al., 2016) Finally, the Good Behavior Game {636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008). In an outcome evaluation of the 686, first graders assigned to 686 reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for 636 to be delivered with precision, consistency, and teacher support. (536 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2003). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self- regulation and social competence (all protective factors for suicide) (Herman et al., 2011}. The program includes 9- 20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8; Hammond, 2003; C. Webster- 25 Stratton Hammond, 1997; C. Webster-Stratton, Reid, 8: Hammond, 2001]. The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionally, Strengthening Families 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guvll, Dav, 2002]. Strengthening Families has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 26 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at- risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. Pepple who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults {and youth}. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t ?slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. 0 Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers and/or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of 27 depression, hopelessness, and subsequent mental health care utilization. Like means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes 0 Reduction in suicide attempts I Reduction in suicide deaths 0 Increased identification of individuals at-risk for suicidal behavior I Increased at?risk individuals in treatment I Increased community members trained to identify at-risk individuals I Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained (Substance Abuse and Mental Health Services Administration, 2014). However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena 31 Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. 0 Gatekeeper training. Mental Health First Aid (MHFA) is a program designed for the lay public, which consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders), possible risk factors, and where and how to get evidence-based effective help (Kitchener 3: Jorm, 2004). In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener 8: Jorm, 2004). Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener Jorm, 2006). I Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Core program was the pre-cursor to Zero Suicide, and its overall goal was to 28 eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, &Ahmedani, 2015). Crisis intervention. in an evaluation of the effectiveness of the National Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that participants experienced significant decreases in suicidality over the course of the telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). In another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Appiieo? Suicide intervention Training (ASST), a widely implemented gatekeeper training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in ASIST were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. 29 However, training in did not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013]. 30 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co?worker, or other acquaintance to suicide lPitman, Osborn, King, 3: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer 8L Sonneck, 1993; Niederkrotenthaler 81 Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post?discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. 0 Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities le.g., home visits, mail, telephone, e-mail} to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296-2594: reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) 0 Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas Kutcher, 2011}. 31 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public le.g. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andj'or suicide contagion. Potential Outcomes I Reduction in mental health-related sequelae 9 Increase - Improved coping skills 0 Improved messaging following suicide I Reduction in re-attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects ofspecific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation ofsuicide-related mortality requires large sample sizes and extended follow-up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at-risk of suicide. Improving Mood -- Promoting Access to Collaborative Treatment is a program that aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. IMPACT facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase} by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2006; Unutzer et al., 2006) relative to patients who received care as usual. Collaborative Assessment and Management of Suicidality (CAMS) is a therapeutic framework for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient- 32 specific treatment plans. CAMS sessions are collaborative and involve constant patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies {Jobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. (Comtois et al., 2011i Dialectical Behavioral Therapy (DBT) is a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of DBT include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. in a randomized controlled trial of women with recent suicidal or self-injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006]. Attachment-Based Family Therapy (ABFT) is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care (-4.37 vs. -2.34; .001; Additionally, a higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25] and at 24 weeks {82.1% vs. 46.2%; .006; (Diamond at al., 2010). Treatment to prevent re-attempts. Emergency Department Brief Intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months {at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths 33 from suicide relative to a treatment-as-usual group versus respectively; chi2 13.83, 0.001) (Fleischmann et al., 2008). Active follow-up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years). These approaches have been found in a meta-analysis conducted by lnagaki et al. (2015) to reduce reattempts by approximately 17% for up to 12 months post- discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta?analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom {2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian- Moghaddam, Sarjami, Kolahi, Carter, 2011; Wang et al., 2016). Cognitive Behavior Therapy for Suicide Prevention is a strategy that uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CST- SP also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial found of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual {Brown et al., 2005). Postvention programs such as StondBy Response Service are implemented with the goal of providing support to survivors of suicide. StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, 8: Scuffham, 2014]. In a study by Visser et al. (2014}, StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program {48% and 64% 34 respectively, 0.005). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self- refer for services) (J. Cerel 8; Campbell, 2008). Safe messaging following a suicide. Media guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline} and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guideiines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually {95% confidence interval: -149 to -2.32, df 54, 0.024} in the Viennese subway system {Niederkrotenthaler 8: Sonneck, 2007) 35 Sector Involvement Public health can piay an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business/la bor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016). The National Electronic Injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.), age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose {Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts It is also important at all levels ilocal, state, and federal} to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk 37 and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community?level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release ofthe first world report on suicide, more timely surveillance data, and critical mention in the President?s FY17 budget, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking), and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination-?in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re- 38 attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantlv, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer levels of the social ecology, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the smence. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 39 References {See last 7 pages after Appendix?EndNete put them there because it thought the Appendixftable was part of the regular text; Didn?t want to cut and paste table for fear it?d get tie-formatted) 4O Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen ?nancial security Best Available Evidence Suicide Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Strengthen Unemployment bene?t programs Government Economic (local, state, supports Federal) Housing stabilization policies The National Neighborhood Stabilization Government Program (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parityr Laws Healthcare care Government (local, state, Federal) Means restriction intervening at hot spots Government (local, state, Establish 5 5 Eedera'l protective afe storage practices (attempts) a storage overnment . of ?rearms and (local, state, environments . . medication} Federal) Organizational policies and culture Best Availabl Evidence Togetherfor Life Business/Labor US Air Force Suicide Prevention Program (family Government violence) (local, state, Federal) Business/Labor Community?based policies to reduce excessive alcohol use Aicohoi outiet density Government (local, state, Federal) Peer norm approaches Promote connectedness SDWCES of Strength Public Health to protect Social Services against suicide actwatles Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Manta} Heaith Program Public Health Social Services Signs of Suicide Teach co in 3 Good Behavior Game and problem- solving skills Parenting skill and family relationship approaches The incredibie Years Public Health Social Services Strengthening 10-14 Best Availabl Evidence Gatekeeper training Mentai Heaith First Aid Public Health Healthcare Social Services Identify and Screening combined with care management people Henry Ford Perfect Depression Care {Pre- Healthcare cursor to Zero Suicide) Crisis Intervention Nationai Suicide Prevention Lifeiine Public Health Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide improving Mood - Promoting Access to Healthcare Coiiaborative Treatment Social Services Intervene to Ufogcigg?gtrigjggessment and Management lessen harms and We?? Diaiecticai Behaviorai Therapy future risk Attachment-Based Famiiy Therapy Treatment to prevent rerattempts Best Availabl Evidence ED Brief intervention with Follow-up Visits Health ca re Active follow-up con tact approaches EST for Suicide Prevention Postvention Stono?By Response Service Healthcare Safe messaging following a suicide Medio Guidelines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as nongovernmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. REFERENCES Arsenault-Lapierre, (3., Kim, C., 8.: Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37'. doi: 10.1186f1471-244X-4-37 Ba ndura, A. (1986). Socioifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), 336-347. Basile, K. C., Hamburger, M. E., Swahn, M. H., 8; Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14(4), 329. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust i's'ZJt 35. doi: 10.1080f0004862010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., 8: Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust 43. doi: Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi: Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i6?i11. doi: Borowsky, I. Hogan, M., 84 Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A, MacDonald, J. M., Tam, v. W., Jackson, T. D., 8: Ten Have, T. R. (2011). A difference? in?differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 174(11), 1296?1306. doi: 10.1093iaje/kwr273 Brown, 6. K., Ten Have, T., Henriques, G. R., Xie, S. X., Hollander, J. E., Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?520. doi: 10.1001/jama.294.5.563 Butchart, A., Phinney, A., Check, P., 8; Villaveces, A. (2004). Preventing Vioience:A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M., Knoble, N. B., Shortt, J. W., 81 Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231-280. doi: 10.1891f1946?656032.231 Carter, M, McGee, R., Taylor, 8., 3: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51-62. doi: Cavanagh, J. T., Carson, A. J., Sharpe, M., St Lawrie, S. M. (2003). autopsy studies of suicide: a systematic review. Med, 33(3), 395-405. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. from Available at: Strategic Direction Full version-a.pdf Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016:). Sexual Violence: Risk and Protective Factors. from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2015e). Youth Violence: Risk and Protective Factors. from Available at: Cerel,1., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. doi: 10.1521isuli.2008.38.1.30 Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of "Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lhfe-threatening behavior, 44(6), 591-600. doi: Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Threat Behav, 37(5), 543-552. doi: 10.1521/suli.2007.37.5.543 Cherpitel, C. 1., Borges, G. L. G., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-285. Classen, T. 1., 84 Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heuith Econ, 21(3), 338-350. doi: 10.1002fhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524-1526. doi: 10.1176ips.2006.5?.10.1524 Coffey, C. E., Coffey, M. 1., 8i. Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. doi: 10.1176/appi.PS.640422 Coffey, M, Coffey, C., 84 Ahmedani, B. K. (2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. doi: Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963372. doi: 10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M, Spilsbury, J. C., 8: Korbin, J. E. How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negl', 111241142. doi: Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. doi: Crosby, A. E., Han, I3., Ortega, L., Parks, 5. E., 84 Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 55-13), 1-22. Crosby, A. E., Ortega, L., 8.: Melanson, C. (2011}. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Crosby, A. E., 8: Sacks, 1. 1. {2002). Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav, 32(3), 321-328. Curtin, S. C., Warner, M., 8: Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., 8-1. Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi: 10.1093/aje/kwu106 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, 1. A. Mercy, A. B. Zwi, at R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 1-56}. Geneva, Switzerland: World Health Organization. DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, l. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. (2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. AmiI 162(2), 311-318. doi: 10.1176/appi.ajp.162.2.311 Diamond, G. 5., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 8: Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 3; G. Simpson, trans). Glencae, it: Free Press.{Original work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 84. Vella-Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 3? countries. lAdolesc Health, 45(4), 351-359. doi: Escobedo, L. 6., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 689694. Etzersdorfer, E., Sonneck, G. (1998). Preventing Suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. 0., Hindman 1., Reed, G. A., 8: Iskander, J. {In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 3: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195209. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi: 50042~ 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., Frazier, L. (2015}. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311-316. doi: 10.21051AJ PH.2014.301945 Freisthler, B., Merritt, D. H., 8i. LaScala, E. A. (2006). Understanding the ecology ofchild maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263280. doi: 10.117211077559506289524 Frieden, T. R. (2010). A framework for public health action: the health impact pyramid. Am Public Health, 100(4), 590-595. doi: 10.21051AJPH.2009.185652 Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am 1 Public Health, 104(1), 17-22. doi: Ghahramanlou-Holloway, M., Bhar, 5., Brown, 6., Olsen, (2., 8; Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. doi: 10.1111fadd.12762 Goldsmith, 5. K., Pellmar, T. C., Kleinman, A. M., 31. Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. H., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(5), Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. doi: 10.1521fsu i.2007.37.3.333 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-214. doi: 10.1001fjama.293.6.707 Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Harris, K. M., Carpenter, C., 8: Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. doi: Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8L Carter, 6. L. {2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Journai of 198(4), 309-316. Hawton, K. {2007}. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi: Heise, L., 8: Garcia-Moreno, C. {2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, R. Lozano {Eds}, Worid Report on Vioience and Heaith (pp. 87-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M, 3: Webster?Stratton, C. (2011}. The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's (Io?Occurring Internalizing (2, 26(3), 189?201. doi: 10.1037fa0025228 Hong, J. 5., Kral, M. J., Espelage, D. L., St Allen?Mea res, P. {2012). The social ecology of adolescent?initiated parent abuse: 3 review of the literature. Chiid Hum Dev, 43(3), 431?454. doi: 10.1002/510528? 011?02234; Hunkeler, E. M, Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259?263. doi: 10.1136fbmj.38683.210255.BE Inagaki, M, Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a metamanalysis. JAffect Disord, 175, 66?73. doi: 10.1016fj.jad.2014.12.048 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence? based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi: 10.1111lj.1943- 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 81 Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124*2133. doi: Kellam, S. 6., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohoi Depend, 95 Suppi 1, 55-528. doi: Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. doi: 10.1186f1471-244X-4-23 Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust ZJ 40(1), 6-8. doi: Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. {2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi: 10.1022/0227- 59101a000136 Klevens, J., Simon, T. R., 81 Chen, J. (2012). Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. interpers Violence, 27(10), 1987-2002. doi: 10.1177i0886260511431441 Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 32212428), 1326. Knox, K. L., Pflanz, S., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457-2463. doi: 10.2105fajph.2009.159871 Knox, M. S., Burkha rt, K., St Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 23-88. doi: 10.1002lhec.1816 Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. doi: Losel, F., Farrington, D. P. [2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58?523. doi: 10.1016/j.amepre.2012.04.029 Luo, F., Florence, C. 5., QuispewAgnoli, M, Ouyang, L., 81 Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928?2002. Am Public Health, 101(6), 1139?1146. doi: 10.2105iajph.2010.300010 Maimon, D., Browning, C. R., 81 Brooks?Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307?324. doi: 10.11721'0022146510372878 Mishara, B. L. (1995). The impact ofSuicia?e. New York: Springer. Mishara, B. L., 8t Martin, N. (2012}. Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162468. Mitchell, A. M, Kim, Y., Prigerson, H. G., 8; Mortimer?Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, Motto, J. A., Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828?833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. {2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. doi: National Action Alliance for Suicide Prevention. (2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention. (2016). Action Alliance Priorities. Retrieved July 12, 2016, from National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust ZJ 41. doi: 10.1080i00048670701266680 Olfson, M., Gerhard, T., Huang, C., 5., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 22(12), 1172-1181. doi: Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193?199. Pena, J. B., 81 Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614-632. doi: 10.1521f5u i.2006.36.6.514 Perron, S., Burrows, S., Fournier, M., Perron, P. A., 8: Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Heaith, 103(7), 1235-1239. doi: 10.2105!ajph.2012.301089 Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma 1Lr?ioi'ence Abuse, 13(2), 112-132. doi: 10.1172/15248380124456441 Pirkis, J., Too, L. 5., Spittal, M. J., Krysinska, K., Robinson, J., 8: Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. doi: Pitman, A., Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. R., 3: Williams, J. M. {2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, D., Chang, S. 5., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. doi: 2 Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 411?491. Resnick, M. 0., Ireland, M., 8: Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdoiesc Heaith, 35(5), 424 e421? Rowhani?Ra hba r, A., Simonetti, J. A., 8: Rivara, F. P. (2016). Effectiveness of interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111?124. doi: Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., 81 Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med, 17(1), 8?14. doi: 10.5811iwestjem.2015.11.28590 Runyan, D., Wattam, C., Ikeda, R., Hassan, F., 81 Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, at R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., Gannon?Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annuai Review ofSocioiogy, 28, 443428. Schilling, E. A., Aseltine, R. H., Jr., 8; James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 12(2), 152?166. doi: Shepard, D. 5., Gurewich, D., Lwin, A. K., Reed, G. A., Jr., 81 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. doi:10.1111isltb.12225 Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and iife- threatening behavior, 25(1), 10-21. Simon, 0. R., Swarm, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., Day, S. X. (2002). Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. Stud Aicohoi, 63(2), 219- 223. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., 8; Carpenter, M. {2008). Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136-142. doi: 10.1176/appi.ap.32.2.136 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Public Health, 102(1), 13-29. Townsend, E., Hawton, K., Altman, D. (3., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979-933. U.S. Office of the Surgeon General, 3: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. 1.4m Geriatr Soc, 54(10), 1550-1556. Visser, V. 5., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. lournalof Community 42(1), 19-23. doi: 10.1002/jcop.21536 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. doi: 10.1111/sltb.12177 Wasserman, D., Haven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster?randomised, controlled trial. The Lancet, 385(9977), 1536?1544. doi: Webster?Stratton, C., 81 Hammond, M. Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consult Clin 65(1), 93?109. Webster?Stratton, (2., Reid, M. J., Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 2333302. doi: Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child 3 Adolescent 40(2), 191003. Webster?Stratton, C., Jamila Reid, M., Stoolmiller, M. (2003). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in highsrisk schools. Journal of child and 49(5), 471?433. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2003). Measuring neighborhood connection and the association with violence in young adolescents. JAdolesc Health, 43(5), 432439. doi: Wilcox, H. C., Kellam, S. (3., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., 8: Anthony, J. C. (2003). The impact of two universal randomized first- and classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?73. Wilson, W. J. (2011). When warlr disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A., Brown, C. H., LoMurray, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8L Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. doi: relationships? Keep the levels the same as Comment Should we acid the SEM or is this ok? I i Policies. Programs, and Practices to Support Individuals, [Familiesi 8: Communities: A Technical Package to Prevent Suicide Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow. Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for injury Prevention and Control (NCIPC) Centers for Disease Control and Prevention 2016 Poiicies, Programs, and Practices to Support Individuals, Communities: A Technicai Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercv, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, EL, Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technical Package to Prevent Suicide. Atlanta, GA: National Center for Injurv Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements External Reviewers . Overview of Technical Package Strengthen Economic . . . . .. . . Strengthen Access to Mental Health Care Create Protective Environments Promote Connectedness Teach Coping and Problem?Solving Skills Identify and Support People At Risk Intervene to Lessen Harms and Prevent Future Risk Sector involvement Monitoring and Evaluation Conclusion References Appendix External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Please include Helen Singer, MPH Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness to protect against suicide; teaching caping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation ofthis package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8i Melanson, 2011). Self-directed violence may be saicidoi or non-suicidoi in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. it contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8: Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 13 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicide (Le. ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model-- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community level: Inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). lt is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8: Villaveces, 2004; Klevens, Simon, Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, lrwin, Spilsbury, Korbin, 2007; Freisthler, Merritt, 8; LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8.1 Kim, 2012), intimate partner violence (Pinchevsky Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 8a. Ramiro, 2002), intimate partner violence (Heise 8c Garcia-Moreno, 2002; Pinchevsky 8c Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 8 2007; Kleiman, Riskind, Schaefer, 8-1 Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8t Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Br Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen-Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8i Vella-Zarb, 2009; Maimon, Browning, Brooks-Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8; Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8a Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8i. Sacks, 2002). Research indicates that the health consequences of violence, including suicide, are also much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt (Chapman Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, BL Carpenter, 2008). The economic toll ofsuicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8r Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (U.S. Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman Maris, 1995; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor] (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. 10 Comment Can we just call this i Promote Connectedness so it?s more i consistent with other strategy titles? Not i sure why we have to say to protect against suicide. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security Housing stabilization policies Strengthen access to mental health care 0 Coverage of mental health conditions in health insurance policies Create protective environments - Reducing access to lethal means among persons at?risk ofsuiclde Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness protect against - Peer norm approaches -. ?Lem its neaserne cease titles suicidd??m Teach coping and problem?solving skills Social?emotional learning Parenting skill and family relationship approaches Gatekeeper training Screening combined with care management Identify and support people at risk in Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re?attempts Intervene to lessen harms and prevent future risk I Postvention Safe messaging following a suicide The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. 11 Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide?related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as businessflabor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working?age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015}. Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation ofsuch financial stress, increase an individual?s riskfor suicide and how buffering these risks can potentially protect against suicide {Stack Wasserman, 2007). Public health research suggests that some of the biggest impacts on health come from efforts directed at improving the socioeconomic conditions of individuals and families {Frieden, 2010}. Strengthening economic support systems can help people stay in their homes or obtain affordable housing, pay for necessities such as food and medical care, and get job training, among other things. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress - Strengthening household financial security can reduce the risk of suicide by providing individuals with the financial means to cover basic expenses and lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits, livable wages, Temporary Assistance to Needy Families (TAN medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability are examples of ways to strengthen household financial security. II Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other Options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reduced suicide rates I Lower foreclosure rates 0 Lower eviction rates 13 0 Reduced emotional distress Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household financial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits iCylus, Glymour, 8: Avendano, 2014}. Another US. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk (Classen Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance) have also shown an impact on suicide. A study by Flavin and Radcliff {2009) examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Fiavin Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied. At the national level, an estimated 3,000 fewer suicides would occur per year nationwide if every state increased their per capita spending on these types of assistance by $45 per year iFlavin 8L Radcliff, 2009}. Housing stabilization policies. The i?iiationail Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working?aged adults iHoule 81. Light, 2014). Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found 14 that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent's home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access to Mental Health Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, Stroup, 2015; Owens, 2002), and risk conferred by mental illnesses differ (Arsenault? Lapierre, Kim, 8: Turecki, 2004; E. C. Harris 81 Barraclough, 1997; Tyrer, Reed, 8: Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 84 Barraclough, 1998; World Health Organization, 2014). Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths {Arsenault?Lapierre et al., 2004; Cavanagh, Carson, Sharpe, 8: Lawrie, 2003; Isometsa, 2001). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8: Cao, 2007). Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions {Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention (World Health Organization, 2014). Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law includes coverage for some mental health conditions but not others). Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes I Increased utilization of mental health services 16 I Decreased of mental illnesses I Decreased rates of suicide attempts - Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. II Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006} found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of $1.3-3.1 million per suicide prevented (Lang, 2013). 17 Create Protective Environments Etatinnale] Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddon, 1980}. Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2015; et al., 2016), therefore, changes to the organizational culture in these occupations, by way of implementation of supportive policies or even physical modi?cations to the workplace environment, can demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide, are not (K. L. Knox et al., 2010: National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, changing the availability of and access to common suicide risk factors in the environment such as bridges, alcohol, medications, or ?rearms, can reduce suicide rates, particularly in times of crisis {Beautrais, Gibb, Fergusson, Horwood, 84 Larkin, 2009; Crosby, Espitia~Hardeman, Ortega, 8.: Lozano, 2013; Kaplan et al., 2013: Miller, Warren, Hemenway, 8t Azrael, 2015; C. W. Runyan et al., 2015]. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. 4- Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanginglsuffocation, or jumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1) the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer at al., 2009; Simon et al., 2001} and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007,- Yip et al., 2012). Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches redUcing access to lethal means for persons at?risk of suicide: Interval/ling otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures leg, bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include 18 Comment IAI: Not sure if the additions i made to the rationale are too duplicative of info below. If so, go ahead and delete. erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition-?and keeping medicines in a locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts lRowhani-Rahbar, Simonetti, Rivara, 2016; C. W. Runyan et al., 2015i - Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (National Action Alliance for Suicide Prevention Workplace Task Force, 2015]. - Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8: Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {Cherpitel, Borges, 8L Wilcox, 2004). Potential Outcomes - Increase in safe storage of means - Reduction in suicide attempts 0 Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2015), as described below. 19 Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal ofthe barrier, sadly, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016]. Another program, The Emergency Department Counseiing on Access to Lethoi Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et. al {2016) found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara St Martin, 2012). Police suicides were tracked over 12 years and 20 compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara En Martin, 2012i Another example of this approach is the United States Air Force Suicide Prevention Program AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8a Caine, 2003}. Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community?based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewaid, Remer, 2009). 21 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002}, for instance between individuals peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one's community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community,?l neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all ofwhich helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 22 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes - Reduction in maladaptive coping attitudes and behaviors 0 Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed I Increase help-seeking behaviors I Positive perception ofadult support Eviden ce Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (6 metropolitan, 12 rural], Wyman et al. (2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism [Branas et al., 2011). 23 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1985), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters (Pollock Williams, 2004}. Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8; Bunney, 2002) and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, 8: Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 24 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 81 Hunter, 2010]. Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program {Wasserman et al., 2014). In a cluster- randomized controlled trial of YAM conducted across 10 European Union countries and 163 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=D.025] at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% (Wasserman et al., 2014). Signs ofSuicide {505) is another school-based prevention program for students aged 13:11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 8: James, 2016}. In a randomized controlled trial, 505 was shown to reduce self-reported suicide attempts at 3-months post intervention among 25 participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression andlor suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling et al., 2016). Finally, the Good Behavior Game (636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the GBG is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2003). In an outcome evaluation of the 656, first graders assigned to GBG reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (386 effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between (336 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for (536 to be delivered with precision, consistency, and teacher support. 656 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years NY) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide) (Herman et al., 2011). The program includes 9- 20 sessions offered in communitynbased settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, lamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8; Hammond, 26 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionallv, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 27 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims ofviolence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults (and youth). I Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t 'slip through the cracks?. These approaches typically employ screening for depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotiine, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 28 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at-risk individuals in treatment I Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained {Substance Abuse and Mental Health Services Administration, 2014]. However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena 8i. Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. II Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid (MHFA) program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders}, possible riskfactors, and where and how to get evidence-based effective help (Kitchener Jorm, 2004). In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener Jorm, 2006). Gatekeeper training has also been a core part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the 29 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GLS trainings had significantly lower youth suicide rates the year following the training implementation (-1.07, p=.03) (Walrath, Garraza, Reid, Goldston, 8: McKeon, 2015) This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 2007 and 2010. Counties implementing GL3 program activities also had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation of the GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% CI, 1.8-8.0 fewer attempts per 1000 youths]; .003; (Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015)}. More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GLS program. Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Core program was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006]. An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8a. Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 8: Ahmedani, 2015). Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also 30 found that participants experienced significant decreases in suicidality over the course of the telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). In another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Applied Suicide intervention Training (ASST), a widelyr implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safety connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful bv the end of their call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 31 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide (Pitman, Osborn, King, 8: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer E: Sonneck, 1998; Niederkrotenthaler 81 Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%_25% reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas 8: Kutcher, 2011}. 32 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andfor suicide contagion. Comment IAI: The Write?up of the evidence for this strategy needs to follow the format used in the other sections of the TP. I modified the write-up for the first approach to illustrate the form at. You will need to do something similar for the second approach. Potential Outcomes Reduction in mental health?related sequelae . Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow~up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of Impact on risk and protective factors for suicide. One example is the Improving Mood? Promoting Access to Coiiaborotive Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. it facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et ai., 2006}. The program has been shown to significantly improve quality.r of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2005; Unutzer et al., 2006?; relative to patients who received care as usual. Another example is Collaborative Assessment and Management ofSuicidaiity which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 33 patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (lobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. {Comtois et al., 2011). Other examples include Diaiecticol Behoviorai Therapy (DST) and Attachment-Based Family Therapy (ABFT). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two?year follow-up than women receiving community treatment {23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined (Linehan et al., 2006]. ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety [Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care {-4.37 vs. -2.34; .001; Additionally, a higher percentage participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (32.1% vs. 46.2%; .006; {Diamond et al., 2010). Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts among those at risk for suicide have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief intervention with Foiiaw-Up Visits is a program that involves a one? hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months {at 1, 2, 4, 7, 11 weeks and 4, 6,12, 18 months]. Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18- months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from 34 suicide relative to a treatment-as-usual group versus respectively; chi2 13.83, --: 0.001) (Fleischmann et al., 2008). Another example of treatment to prevent re-attempts involves active follow-up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years). These approaches have been found in a meta-analysis conducted by Inagaki et al. {2015) to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001} found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial found of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual [Brown et al., 2005). Postvention programs are implemented with the goal of providing support to survivors of suicide to reduce their own risk of suicide. One example of a postvention program, StandBy Response Service {StandBy}, provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Coma ns, 35 Scuffha m, 2014). In a study by Visser et al. (2014), StondBv clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (43% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) U. Cerel Campbell, 2008). Safe messaging following a suicide. One way to ensure safe messaging following a suicide is to encourage that reporters adhere to media guidelines for reporting on suicides. Such guidelines can help assure that stories on suicide are communicated in an effective way that reduces risk to others who may be vulnerable. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually {95% confidence interval: -149 to ?13; -2.32, df 54, 0.024} in the Viennese subway system {Niederkrotenthaler 8: Sonneck, 200?] 36 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. 37 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System?All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 38 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, more timely surveillance data, and critical mention in the President?s FY17 budget, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking), and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination?-in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re-attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer 39 levels of the social ecologv, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 40 References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 41 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2016a). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel,1., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8; Marshall, D. (2014). The Continuum of ?Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lace-threatening behavior, 44(6), 591-600. Chapman, A. L., S: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-235. Classen, T. 1., 8: Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi:10.1002lhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524?1526. Coffey, C. E., Coffey, M. 1., 3a Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M, Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28{11), 963072. doi:10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M., Spilsbury, 1. C., 8: Korbin, 1. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negi, 11111142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 81'. Loza no, B. (2013). Alcohol and suicide. Aicohoi; Science, Poiicy and Pubiic Health, 190. Crosby, A. E., Han, 0., Ortega, L., Parks, 5. E., 84 Gfroerer,1. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 1:13 Years United States, 2003*2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directeo' Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Cunningham, P. 1. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff {Miilwood}, 28(3), w490-501. Curtin, S. C., Warner, M., 8L Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., Avendano, M. {2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am 1 Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 42 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on lv?iolence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, 5., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Arnl 152(2), 311-318. doi:10.1176lappi.ajp.162.2.311 Diamond, (3. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. [2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987'). Suicide aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8L G. Simpson, trans). Glencoe, it: Free work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 3i. \lella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adolesc Health, 45(4), 351?359. Escobedo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accia' Anal Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, 6. A., 8: lskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., \lagi, K. J., Barnes, J., El. Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311?316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263-280. doi:10.1177l1077559506289524 Frieden, T. R. (2010). A framework for public health action: the health impact pyramid. Am Public Health, 100(4), 590-595. doi:10.2105lAJ PH.2009.135652 Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, C., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06}, 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 43 Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M, Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. 0., Villaveces, A., Prodzinski, J., . . . Ha rruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Health Reports, 95(5), 411-421. Han, H., Crosby, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., St Gfroerer, J. (2016}. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176-186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205228. Harris, E. C., Barraclough, B. (1998). Excess mortality of mental disorder. 173, 11?53. Harris, K. M, Carpenter, C., 8: Baa, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., at Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce Suicidal behaviours 12 months after hospital?treated self~poisoning. British Journal of 198(4), 309?316. Hawton, K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi:101027702226910.28514 Heise, L., Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano (Eds), World Report on Violence and Health (pp. 82-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 3: Webster~Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q, 26(3), 189-201. Hang, J. S., Kral, M. J., Espelage, D. L., at. Allen?Mea res, P. (2012). The social ecology of adolescentainitiated parent abuse: a review of the literature. Child Hum Dev, 43(3), 431?454. 011-02734; Houle, J. N., Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Public Health, 104(6), 1073-1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-263. lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .iAffect Disord, 1.75, 66-78. Isometsa, E. T. (2001). autopsy studies--a review. for 16(7), 379-385. 44 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 84 Flemer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12), 2124-2133. Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, H., Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. Prev, 19(1), 38-43. Kellam, S. (3., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. (3., leson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. hi EnglJ Med, 352(24), 2515-2523. Kitchener, B. A., Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust ZJ 40(1), 6?8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 8: Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273?279. 5910/a000136 Klevens, J., Simon, T. R., at Chen, J. {2012). Are the perpetrators of violence one and the same? Exploring the co? occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 27(10), 1937?2002. dDi:10.117if0386260511431441 Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 32347423), 1326. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 24512463. Knox, M. S., Burkha rt, K., 81 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment?related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 73-88. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(2), 752-266. Losel, F., 8; Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. 5., Quispe-Agnoli, M., Ouyang, L., 8: Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am Public Health, 101(6), 1139-1146. doi:10.2105fajph.2010.300010 Maimon, D., Browning, C. R., Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Sherri, D. L., Bagalman, J. E., 8; Cao, Z. (2007'). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 45 McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortai Wkiy Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., at Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., 8: Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journai of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8a Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., St Zhu, W. {2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action pian to save iives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Biueprintfor Workpiace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and weiiness with insights from iived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: 0.5. Nationai Guideiines. Washington, D.C. Author. Niederkrotenthaler, T., 81 Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust 21 41. Olfson, M., Gerhard, T., Huang, (2., 5., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172?1181. Owens, D. (2002). Fatal and non-fatal repetition of selfsharm: Systematic review. British Journai of 181(3), 193?199. Pena, J. B., St Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614?632. Perron, 5., Burrows, 5., Fournier, M., Perron, P. A., B: Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. ?lmsf Pubiic Health, 103(7), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Vioience Abuse, 13(2), 112?132. Pirkis, J., Too, L. S., Spittal, M. J., Krysinska, K., Robinson, J., St Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-16?. Reeves, A., Stuckler, D., McKee, M., Gunnell, D., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. 46 Resnick, M. D., Ireland, M., Borowsky, I. (20041). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. Adolesc Health, 35(5), 424 e421- 410. Rowhani-Rahba r, A., Simonetti, J. A., Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiol Rev, 38(1), 111-124. Ru nyan, C. W., Becker, A., Brandspigel, 3., Barber, (2., Trudeau, A., 81. Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med, 17(1), 3-14. Runyan, D., Wattam, C., lkeda, R., Hassan, F., 8: Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano (Eds), World Report on 'v'iolence and Health (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., 8! Gannon-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review ofSocialogy, 28, 443-473. Schilling, E. A., Aseltine, R. H., Jr., 8t James, A. [2016). The 505 Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 1212), 152-166. Shepard, D. S., i3urewich, D., Lwin, A. K., Reed, 6. A., Jr., 3t Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and life? threatening behavior, 25(1), 10?2 1. Simon, O. H., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 8i O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M, 81. Day, 5. X. (2002). Universal family?focused interventions in alcoholvuse disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 228. Stack, 5., 3t Wasserman, l. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 32(1), 103?112. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., 31. Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136?142. doi210.1176/appi.ap.32.2.136 Szumilas, M., Kutcher, S. (2011). Post?suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. (3., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: metaea nalysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 929?988. Tyrer, P., Reed, G. M., 8: Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 81 Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 42(1), 19-28. 47 Walrath, C., Ga rraza, L. (3., Reid, H., Goldston, D. B., 8-: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105iAJPH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 1536-1544. Webster-Stratton, C., 8; Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin 55(1), 93-109. Webster-Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 233-302. Webster-Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for 1young children with ADHD. Journal of Clinical Child 8: Adolescent 40(2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schools. Journal of child and 49(5), 471?488. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2008). Measuring neighborhood connection and the association with violence in young adolescents. Adolesc Health, 43(5), 482?489. Wilcox, H. C., Kellam, S. 6., Brown, C. H., Poduska, l. M., Ialongo, N. S., Wang, W., &Anthonv, J. C. (2008). The impact of two universal randomized first? and second?grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?23. Wilson, W. J. (2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A., Brown, C. H., LoMurrav, lvl., Schmeelk?Cone, K., Petrova, M, Yu, (1., . . . Wang, W. (2010). An outcome evaluation ofthe Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653?1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8; Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 48 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Approach/Program, Practice or Policy Suicide Suicide Other Lead Sectors1 Attempts or Riskarotective ldeation Factors for Suicide Strengthen household financial security Strengthen Unemployment bene?t programs 1/ Government economic (local, state, supports Federal) Housing stabilization policies The National Neighborhood Stabilization 1/ Government Program (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parityr Laws v/ v? Government care (state, Federal) Health care Reducing access to lethal means among persons at?risk intervening at hot spots Government (local, state, Establish 5 :eiira: protective afe storage practices 1/ u: ea environments Organizational policies and culture Togetherfor Life 1? Businesleabor US Air Force Suicide Prevention Program Best Available Ev' - Government (local, state, Federal) Businesstabor Community?based policies to reduce excessive alcohol use against suicide Alcohol outiet density v? Government (local, state, Federal) Peer norm approaches Promote connectedness Sources 0f Strength Public Health to protect Social Services Community-engagement activities Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Mental Health Program Public Health Signs of Suicide Education Good Behavior Game solving skills Parenting skill and family relationship approaches The incredible Years Public Health Education Strengthening 10-14 Best Available Evidence identify and Gatekeeper training Mental Health First Aid Public Health Healthcare Social Services Screening combined with care management support people Henry Ford Perfect Depression Core (Pre- Health ca re at-risk cursor to Zero Suicide) Crisis Intervention Notionoi Suicide Prevention Lifeiine v? Public Health Social Services Appiieo' Suicide intervention Training Treatment for people at risk of suicide improving Mood Promoting Access to in? Healthcare Coiioborotive Treatment Social Services Coiioborotive Assessment and Management of Suicidoiity CA M3) -st Available Evidence Interue no to Dialectical Behaviorai Therapy 1/ v" lessen harms and prevent Attachment-Based Famiiy Therapy 1/ future risk Treatment to prevent re-attempts ED Brief interven tion with Follow-up Visits 1/ Health ca re Active folio w?up con tact approaches 1/ 1/ CBTfor Suicide Prevention Postvention 5 tandBy Response Service 1" Health ca re Safe messaging following a suicide Media Guideiines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. relationships? Keep the levels the same as Comment Should we acid the SEM or is this ok? I i Policies. Programs, and Practices to Support Individuals, [Familiesi 8: Communities: A Technical Package to Prevent Suicide Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow. Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for injury Prevention and Control (NCIPC) Centers for Disease Control and Prevention 2016 Poiicies, Programs, and Practices to Support Individuals, Communities: A Technicai Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercv, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, EL, Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technical Package to Prevent Suicide. Atlanta, GA: National Center for Injurv Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements External Reviewers . Overview of Technical Package Strengthen Economic . . . . .. . . Strengthen Access to Mental Health Care Create Protective Environments Promote Connectedness Teach Coping and Problem?Solving Skills Identify and Support People At Risk Intervene to Lessen Harms and Prevent Future Risk Sector involvement Monitoring and Evaluation Conclusion References Appendix External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Please include Helen Singer, MPH Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness to protect against suicide; teaching caping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation ofthis package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8i Melanson, 2011). Self-directed violence may be saicidoi or non-suicidoi in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. it contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8: Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 13 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicide (Le. ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model-- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community level: Inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). lt is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8: Villaveces, 2004; Klevens, Simon, Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, lrwin, Spilsbury, Korbin, 2007; Freisthler, Merritt, 8; LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8.1 Kim, 2012), intimate partner violence (Pinchevsky Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 8a. Ramiro, 2002), intimate partner violence (Heise 8c Garcia-Moreno, 2002; Pinchevsky 8c Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 8 2007; Kleiman, Riskind, Schaefer, 8-1 Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8t Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Br Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen-Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8i Vella-Zarb, 2009; Maimon, Browning, Brooks-Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8; Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8a Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8i. Sacks, 2002). Research indicates that the health consequences of violence, including suicide, are also much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt (Chapman Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, BL Carpenter, 2008). The economic toll ofsuicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8r Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (U.S. Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman Maris, 1995; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor] (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. 10 Comment Can we just call this i Promote Connectedness so it?s more i consistent with other strategy titles? Not i sure why we have to say to protect against suicide. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security Housing stabilization policies Strengthen access to mental health care 0 Coverage of mental health conditions in health insurance policies Create protective environments - Reducing access to lethal means among persons at?risk ofsuiclde Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness protect against - Peer norm approaches -. ?Lem its neaserne cease titles suicidd??m Teach coping and problem?solving skills Social?emotional learning Parenting skill and family relationship approaches Gatekeeper training Screening combined with care management Identify and support people at risk in Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re?attempts Intervene to lessen harms and prevent future risk I Postvention Safe messaging following a suicide The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. 11 Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide?related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as businessflabor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale [Public health research suggests that some ofthe biggest impacts on health come from policies developed to improve socioeconomic conditions {Frieden, 2010}. Downturns in the economy and increases in unemployment and home foreclosures are associated with increased rates of suicide {Fowlen Gladden, Vagi, Barnes, 8i Frazier, 2015; Luo et al., 2011]. Policies that strengthen economic supports can help people stay in their homes, pay for necessities such as food and medical care, and getjob training, among other things. In providing this support, stress and anxiety and the potential for a crisis situation may be Approaches [Economic and housing supports for individuals and families can be strengthened by improving policies that enhance financial security and stabilize housing for people, especially in times of economic need] I Strengthen household financial security. Etudies from the U5. indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unempioyment rates, particularly for working-age individuals 25 to 54 years old {Luo et al., 2011; 1. ll. Fowler et al., 2015).] Potential {Jutcomes - Reduced suicide rates 0 Lower foreclosure rates a Lower eviction rates 0 Reduced emotional distress Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits lelus, Glymour, 8i Avendano, 2014]. Another U.S. study examining the link between unemployment Comment IAI: The rationale should focus on the strategy - strengthening economic supports - and how and why that might be beneficial in reducing suicide. As part ofthe rationale you can talk about how suicide relates to economic factors historical data shows that suicide rates rise during periods of economic recessions and decreases during periods of economic expansion]: how economic stressors {job loss, long periods of unemployment, reduced income, difficulty covering medical, food, housing expenses, etc] increase risk for suicide and how buffering these risks can potentially JJ?rotect against suicide. You cover this a Comment Statement should be more along the lines of: ?There are a number of approaches that can help strengthen economic supports and buffer against the risk for suicide." _economic stress." Or ?Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of J5. Comment This would be a good introductory sentence to the rationale. For the two approaches, suggest saying something along the following lines: Strengthening household ?nancial security can reduce the risk for suicide by providing individuals with the necessary means to cover basic expenses and lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits, livable wages, Temporary Assistance to Needy Families medical bene?ts, and retirement and disability insurance to help cover the costs of basic necessities or to offset costs and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed tojust the loss of job, predicted suicide risk {Classen 3i Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to 13 strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance) have also shown an impact on suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied. At the national level, an estimated 3,000 fewer suicides would occur per year nationwide if every state increased their per capita spending on these types of assistance by $45 per year (Flavin 8: Radcliff, 2009]. Housing stabilization policies. The Nationall Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working?aged adults {Houle Light, 2014). Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began) to 2010 (after it had peaked; Fowler et al. {2015)}. Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 14 Strengthen Access to Mental Health Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, Stroup, 2015; Owens, 2002), and risk conferred by mental illnesses differ (Arsenault? Lapierre, Kim, 8: Turecki, 2004; E. C. Harris 81 Barraclough, 1997; Tyrer, Reed, 8: Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 84 Barraclough, 1998; World Health Organization, 2014). Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths {Arsenault?Lapierre et al., 2004; Cavanagh, Carson, Sharpe, 8: Lawrie, 2003; Isometsa, 2001). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8: Cao, 2007). Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions {Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention (World Health Organization, 2014). Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law includes coverage for some mental health conditions but not others). Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes I Increased utilization of mental health services 15 I Decreased of mental illnesses I Decreased rates of suicide attempts - Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. II Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006} found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of $1.3-3.1 million per suicide prevented (Lang, 2013). 16 Create Protective Environments Etatinnale] Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddon, 1980}. Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2015; et al., 2016), therefore, changes to the organizational culture in these occupations, by way of implementation of supportive policies or even physical modi?cations to the workplace environment, can demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide, are not (K. L. Knox et al., 2010: National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, changing the availability of and access to common suicide risk factors in the environment such as bridges, alcohol, medications, or ?rearms, can reduce suicide rates, particularly in times of crisis {Beautrais, Gibb, Fergusson, Horwood, 84 Larkin, 2009; Crosby, Espitia~Hardeman, Ortega, 8.: Lozano, 2013; Kaplan et al., 2013: Miller, Warren, Hemenway, 8t Azrael, 2015; C. W. Runyan et al., 2015]. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. 4- Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanginglsuffocation, or jumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1) the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer at al., 2009; Simon et al., 2001} and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007,- Yip et al., 2012). Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches redUcing access to lethal means for persons at?risk of suicide: Interval/ling otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures leg, bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include 1? Comment IAI: Not sure if the additions i made to the rationale are too duplicative of info below. If so, go ahead and delete. erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition-?and keeping medicines in a locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts lRowhani-Rahbar, Simonetti, Rivara, 2016; C. W. Runyan et al., 2015i - Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (National Action Alliance for Suicide Prevention Workplace Task Force, 2015]. - Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8: Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {Cherpitel, Borges, 8L Wilcox, 2004). Potential Outcomes - Increase in safe storage of means - Reduction in suicide attempts 0 Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2015), as described below. 18 Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal ofthe barrier, sadly, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016]. Another program, The Emergency Department Counseiing on Access to Lethoi Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et. al {2016) found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara St Martin, 2012). Police suicides were tracked over 12 years and 19 compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara En Martin, 2012i Another example of this approach is the United States Air Force Suicide Prevention Program AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8a Caine, 2003}. Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community?based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewaid, Remer, 2009). 20 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002}, for instance between individuals peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one's community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community,?l neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all ofwhich helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 21 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes - Reduction in maladaptive coping attitudes and behaviors 0 Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed I Increase help-seeking behaviors I Positive perception ofadult support Eviden ce Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (6 metropolitan, 12 rural], Wyman et al. (2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism [Branas et al., 2011). 22 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1985), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters (Pollock Williams, 2004}. Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8; Bunney, 2002) and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, 8: Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 23 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 81 Hunter, 2010]. Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program {Wasserman et al., 2014). In a cluster- randomized controlled trial of YAM conducted across 10 European Union countries and 163 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=D.025] at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% (Wasserman et al., 2014). Signs ofSuicide {505) is another school-based prevention program for students aged 13:11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 8: James, 2016}. In a randomized controlled trial, 505 was shown to reduce self-reported suicide attempts at 3-months post intervention among 24 participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression andlor suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling et al., 2016). Finally, the Good Behavior Game (636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the GBG is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2003). In an outcome evaluation of the 656, first graders assigned to GBG reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (386 effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between (336 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for (536 to be delivered with precision, consistency, and teacher support. 656 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years NY) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide) (Herman et al., 2011). The program includes 9- 20 sessions offered in communitynbased settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, lamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8; Hammond, 25 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionallv, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 26 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims ofviolence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults (and youth). I Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t 'slip through the cracks?. These approaches typically employ screening for depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotiine, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 27 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at-risk individuals in treatment I Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained {Substance Abuse and Mental Health Services Administration, 2014]. However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena 8i. Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. II Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid (MHFA) program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders}, possible riskfactors, and where and how to get evidence-based effective help (Kitchener Jorm, 2004). In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener Jorm, 2006). Gatekeeper training has also been a core part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the 28 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GLS trainings had significantly lower youth suicide rates the year following the training implementation (-1.07, p=.03) (Walrath, Garraza, Reid, Goldston, 8: McKeon, 2015) This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 2007 and 2010. Counties implementing GL3 program activities also had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation of the GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% CI, 1.8-8.0 fewer attempts per 1000 youths]; .003; (Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015)}. More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GLS program. Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Core program was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006]. An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8a. Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 8: Ahmedani, 2015). Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also 29 found that participants experienced significant decreases in suicidality over the course of the telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). In another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Applied Suicide intervention Training (ASST), a widelyr implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safety connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful bv the end of their call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 30 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide (Pitman, Osborn, King, 8: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer E: Sonneck, 1998; Niederkrotenthaler Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%_25% reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas 8: Kutcher, 2011}. 31 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andfor suicide contagion. Comment IAI: The Write?up of the evidence for this strategy needs to follow the format used in the other sections of the TP. I modified the write-up for the first approach to illustrate the form at. You will need to do something similar for the second approach. Potential Outcomes Reduction in mental health?related sequelae . Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow~up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of Impact on risk and protective factors for suicide. One example is the Improving Mood? Promoting Access to Coiiaborotive Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. it facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et ai., 2006}. The program has been shown to significantly improve quality.r of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2005; Unutzer et al., 2006?; relative to patients who received care as usual. Another example is Collaborative Assessment and Management ofSuicidaiity which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 32 patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (Jobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. {Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy (AFBT). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined (Linehan et al., 2006]. ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety [Diamond at al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care {-4.37 vs. -2.34; .001; Additionally, a higher percentage participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks {82.1% vs. 46.2%; .006; {Diamond et al., 2010). Treatment to prevent re-attempts. Emergency Department Brief Intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths 33 from suicide relative to a treatment-as-usual group versus respectively; chi2 13.83, 0.001} {Fleischmann et al., 2008}. Active follow-up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years}. These approaches have been found in a meta-analysis conducted by Inagaki et al. (2015} to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta?analysis were small, it was not possible to determine the effect of active contact and follow- up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001} found that patients who refused ongoing care but who were randomized to be contacted by letterfourtimes per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam, Sarjami, Kolahi, Carter, 2011; Wang et al., 2016}. Cognitive Behavior Therapy for Suicide Prevention uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and CST-5P also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial found of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005}. Postvention programs such as Stono?By Response Service are implemented with the goal of providing support to survivors of suicide. StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, 8; Scuffham, 2014}. In a study by Visser et al. (2014}, StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in 34 which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) (J. Cerel Campbell, 2008). Safe messaging following a suicide. Media guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact ofthe guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually [95% confidence interval: -149 to -2.32, df 54, 0.024) in the Viennese subway system (Niederkrotenthaler Sonneck, 2007) 35 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. 36 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System?All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 37 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, more timely surveillance data, and critical mention in the President?s FY17 budget, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking), and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination?-in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re-attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer 38 levels of the social ecologv, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 39 References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 40 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2016a). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel,1., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8; Marshall, D. (2014). The Continuum of ?Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lace-threatening behavior, 44(6), 591-600. Chapman, A. L., S: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-235. Classen, T. 1., 8: Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi:10.1002lhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524?1526. Coffey, C. E., Coffey, M. 1., 3a Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M, Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28{11), 963072. doi:10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M., Spilsbury, 1. C., 8: Korbin, 1. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negi, 11111142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 81'. Loza no, B. (2013). Alcohol and suicide. Aicohoi; Science, Poiicy and Pubiic Health, 190. Crosby, A. E., Han, 0., Ortega, L., Parks, 5. E., 84 Gfroerer,1. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 1:13 Years United States, 2003*2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directeo' Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Cunningham, P. 1. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff {Miilwood}, 28(3), w490-501. Curtin, S. C., Warner, M., 8L Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., Avendano, M. {2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am 1 Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 41 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on lv?iolence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, 5., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Arnl 152(2), 311-318. doi:10.1176lappi.ajp.162.2.311 Diamond, (3. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. [2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987'). Suicide aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8L G. Simpson, trans). Glencoe, it: Free work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 3i. \lella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adolesc Health, 45(4), 351?359. Escobedo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accia' Anal Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, 6. A., 8: lskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., \lagi, K. J., Barnes, J., El. Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311?316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263-280. doi:10.1177l1077559506289524 Frieden, T. R. (2010). A framework for public health action: the health impact pyramid. Am Public Health, 100(4), 590-595. doi:10.2105lAJ PH.2009.135652 Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, C., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06}, 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 42 Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M, Bunney, W. E. (2002). Reducing suicide: A national imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. 0., Villaveces, A., Prodzinski, J., . . . Ha rruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Health Reports, 95(5), 411-421. Han, H., Crosby, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., St Gfroerer, J. (2016}. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176-186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205228. Harris, E. C., Barraclough, B. (1998). Excess mortality of mental disorder. 173, 11?53. Harris, K. M, Carpenter, C., 8: Baa, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., at Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce Suicidal behaviours 12 months after hospital?treated self~poisoning. British Journal of 198(4), 309?316. Hawton, K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi:101027702226910.28514 Heise, L., Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano (Eds), World Report on Violence and Health (pp. 82-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 3: Webster~Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q, 26(3), 189-201. Hang, J. S., Kral, M. J., Espelage, D. L., at. Allen?Mea res, P. (2012). The social ecology of adolescentainitiated parent abuse: a review of the literature. Child Hum Dev, 43(3), 431?454. 011-02734; Houle, J. N., Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Public Health, 104(6), 1073-1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-263. lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .iAffect Disord, 1.75, 66-78. Isometsa, E. T. (2001). autopsy studies--a review. for 16(7), 379-385. 43 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 84 Flemer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12), 2124-2133. Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, H., Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. Prev, 19(1), 38-43. Kellam, S. (3., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. (3., leson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. hi EnglJ Med, 352(24), 2515-2523. Kitchener, B. A., Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust ZJ 40(1), 6?8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 8: Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273?279. 5910/a000136 Klevens, J., Simon, T. R., at Chen, J. {2012). Are the perpetrators of violence one and the same? Exploring the co? occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 27(10), 1937?2002. dDi:10.117if0386260511431441 Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 32347423), 1326. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 24512463. Knox, M. S., Burkha rt, K., 81 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment?related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 73-88. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(2), 752-266. Losel, F., 8; Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. 5., Quispe-Agnoli, M., Ouyang, L., 8: Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am Public Health, 101(6), 1139-1146. doi:10.2105fajph.2010.300010 Maimon, D., Browning, C. R., Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Sherri, D. L., Bagalman, J. E., 8; Cao, Z. (2007'). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 44 McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortai Wkiy Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., at Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., 8: Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journai of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8a Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., St Zhu, W. {2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action pian to save iives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Biueprintfor Workpiace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and weiiness with insights from iived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: 0.5. Nationai Guideiines. Washington, D.C. Author. Niederkrotenthaler, T., 81 Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust 21 41. Olfson, M., Gerhard, T., Huang, (2., 5., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172?1181. Owens, D. (2002). Fatal and non-fatal repetition of selfsharm: Systematic review. British Journai of 181(3), 193?199. Pena, J. B., St Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 36(6), 614?632. Perron, 5., Burrows, 5., Fournier, M., Perron, P. A., B: Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. ?lmsf Pubiic Health, 103(7), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Vioience Abuse, 13(2), 112?132. Pirkis, J., Too, L. S., Spittal, M. J., Krysinska, K., Robinson, J., St Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-16?. Reeves, A., Stuckler, D., McKee, M., Gunnell, D., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. 45 Resnick, M. D., Ireland, M., Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. Adolesc Health, 35(5), 424 e421- 410. Rowhani-Rahba r, A., Simonetti, J. A., Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiol Rev, 38(1), 111-124. Ru nyan, C. W., Becker, A., Brandspigel, 3., Barber, C., Trudeau, A., 81. Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med, 17(1), 3-14. Runyan, D., Wattam, C., lkeda, R., Hassan, F., 8: Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano World Report on 'v'iolence and Health (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., 8! Gannon-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review ofSocialogy, 28, 443-473. Schilling, E. A., Aseltine, R. H., Jr., 8t James, A. [2016). The 505 Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 12(2), 152-166. Shepard, D. S., i3urewich, D., Lwin, A. K., Reed, 6. A., Jr., 3t Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and life? threatening behavior, 25(1), 10?2 1. Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 8i O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M., 81. Day, S. X. (2002). Universal family?focused interventions in alcoholvuse disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 228. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., at Carpenter, M. (2008}. Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136442. Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. 6., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self?harm: meta-a nalysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 929?983. Tyrer, P., Reed, G. M., 8: Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 217326. U.S. Office of the Surgeon General, St National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Ga rraza, L. (3., Reid, H., lE3o dston, D. B., 8-1 McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105lAJ PH.2014.302496 46 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang, M. W., Ko, H. C. (2016). Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Haven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The lancer, 385(9977), 1536-1544. Webster-Stratton, C., Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 65(1), 93-109. Webster-Stratton, C., Reid, M. J., 8: Hammond, M. (2001}. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 233-302. Webster-Stratton, C. H., Reid, M. J., 8: Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008}. Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schools. Journal of child and 49(5), 471-483. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2008). Measuring neighborhood connection and the association with violence in young adolescents. J'Adolesc Health, 43(5), 482-489. Wilcox, H. (2., Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., &Anthonv, J. C. (2008). The impact of two universal randomized first? and second~grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560??3. Wilson, W. J. (2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A, Brown, C. H., LoMurrav, M., Schmeelk?Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 16534661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, Wu, K., at Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 47 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Approach/Program, Practice or Policy Suicide Suicide Other Lead Sectors1 Attempts or Riskarotective ldeation Factors for Suicide Strengthen household financial security Strengthen Unemployment bene?t programs 1/ Government economic (local, state, supports Federal) Housing stabilization policies The National Neighborhood Stabilization 1/ Government Program (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parityr Laws v/ v? Government care (state, Federal) Health care Reducing access to lethal means among persons at?risk intervening at hot spots Government (local, state, Establish 5 :eiira: protective afe storage practices 1/ u: ea environments Organizational policies and culture Togetherfor Life 1? Businesleabor US Air Force Suicide Prevention Program Best Available Ev' - Government (local, state, Federal) Businesstabor Community?based policies to reduce excessive alcohol use against suicide Alcohol outiet density v? Government (local, state, Federal) Peer norm approaches Promote connectedness Sources 0f Strength Public Health to protect Social Services Community-engagement activities Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Mental Health Program Public Health Signs of Suicide Education Good Behavior Game solving skills Parenting skill and family relationship approaches The incredible Years Public Health Education Strengthening 10-14 Best Available Evidence identify and Gatekeeper training Mental Health First Aid Public Health Healthcare Social Services Screening combined with care management support people Henry Ford Perfect Depression Core (Pre- Health ca re at-risk cursor to Zero Suicide) Crisis Intervention Notionoi Suicide Prevention Lifeiine v? Public Health Social Services Appiieo' Suicide intervention Training Treatment for people at risk of suicide improving Mood Promoting Access to in? Healthcare Coiioborotive Treatment Social Services Coiioborotive Assessment and Management of Suicidoiity CA M3) -st Available Evidence Interue no to Dialectical Behaviorai Therapy 1/ v" lessen harms and prevent Attachment-Based Famiiy Therapy 1/ future risk Treatment to prevent re-attempts ED Brief interven tion with Follow-up Visits 1/ Health ca re Active folio w?up con tact approaches 1/ 1/ CBTfor Suicide Prevention Postvention 5 tandBy Response Service 1" Health ca re Safe messaging following a suicide Media Guideiines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. relationships? Keep the levels the same as Comment One idea. Should we add the SEM or is this ok? I i Policies. Programs, and Practices to Support I Individuals, [Familiesi 8: Communities: A Technical Package to Prevent Suicide Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow. Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for injury Prevention and Control (NCIPC) Centers for Disease Control and Prevention 2016 Poiicies, Programs, and Practices to Support Individuals, Communities: A Technicai Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercv, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, EL, Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technical Package to Prevent Suicide. Atlanta, GA: National Center for Injurv Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents cknowledgement External Reviewers Overview ofTechnical Strengthen Economic . Strengthen Access to Mental Health Care .. Create Protective Env1ronment5 Promote Connectedness Teach Coping and Problem?Solving Identify and Support People At Intervene to Lessen Harms and Prevent Future Sector Involvement Monitoring and Evaluation Conelusmn References Comment IAI: I didn?t add the page I numbers vet figuring it might change but I can if you like. Also, Ijust noticed there are two page 5's and no page 4! External Reviewers [to be inserted later] Acknowledgments [to be inserted later] Please include Helen Singer, MPH Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness to protect against suicide; teaching caping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation ofthis package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Frieden, 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8i Melanson, 2011). Self-directed violence may be saicidoi or non-suicidoi in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. it contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8: Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 13 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicide (Le. ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model-- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community level: Inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). lt is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8: Villaveces, 2004; Klevens, Simon, Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, lrwin, Spilsbury, Korbin, 2007; Freisthler, Merritt, 8; LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8.1 Kim, 2012), intimate partner violence (Pinchevsky Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 8a. Ramiro, 2002), intimate partner violence (Heise 8c Garcia-Moreno, 2002; Pinchevsky 8c Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 8 2007; Kleiman, Riskind, Schaefer, 8-1 Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8t Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Br Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen-Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8i Vella-Zarb, 2009; Maimon, Browning, Brooks-Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8; Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8a Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8i. Sacks, 2002). Research indicates that the health consequences of violence, including suicide, are also much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt (Chapman Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, BL Carpenter, 2008). The economic toll ofsuicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8r Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (U.S. Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman Maris, 1995; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor] (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. 10 Comment Can we just call this i Promote Connectedness so it?s more i consistent with other strategy titles? Seems like ?to protect against suicide? is implied in all of the strategies. I Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such ,l that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons i I 4 who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security Housing stabilization policies Strengthen access to mental health care 0 Coverage of mental health conditions in health insurance policies Create protective environments - Reducing access to lethal means among persons at?risk ofsuiclde Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness to protect against - Peer norm approaches -. ?Lem its neaeerne tease titties suicidd??m Teach coping and problem?solving skills Social?emotional learning Parenting skill and family relationship approaches Gatekeeper training Screening combined with care management Identify and support people at risk it Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re?attempts Intervene to lessen harms and prevent future risk I Postvention Safe messaging following a suicide The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. 11 Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide?related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as businessflabor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015]. Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress, increase an individual's risk for suicide; buffering these risks can therefore, potentially protect against suicide [Stack 8t Wasserman, 200?]. For example, strengthening economic support systems can help people pay stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care,job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Approaches Economic sopports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. individuals with the financial means to cover basic expenses and lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits, livable wages, Temporary Assistance to Needy Families (TANFJ, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability are examples of ways to strengthen household financial security. Housing stabilization policies aim to keep people in their homes and provide housing options for Comment Linda, do you think we should say can theoretically reduce the question for housing. I those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes Reduced suicide rates Lower foreclosure rates Lower eviction rates Reduced emotional distress 13 Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household financial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits (Cylus, Glymour, 8i. Avendano, 2014). Another U.S. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk (Classen 8: Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance) have also shown an impact on suicide. A study by Flavin and Radcliff {2009) examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin 81 Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, an estimated 3,000 fewer suicides would occur per year nationwide if every state increased their per capita spending on these types of assistance by $45 per year {Flavin 8: Radcliff, 2009}. Housing stabilization policies. The National Neighborhood Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle 8: Light, 2014). Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 14 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access to Mental Health Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, Stroup, 2015; Owens, 2002), and risk conferred by mental illnesses differ (Arsenault? Lapierre, Kim, 8: Turecki, 2004; E. C. Harris 81 Barraclough, 1997; Tyrer, Reed, 8: Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 84 Barraclough, 1998; World Health Organization, 2014). Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths {Arsenault?Lapierre et al., 2004; Cavanagh, Carson, Sharpe, 8: Lawrie, 2003; Isometsa, 2001). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8: Cao, 2007). Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions {Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention (World Health Organization, 2014). Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law includes coverage for some mental health conditions but not others). Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes I Increased utilization of mental health services 16 I Decreased of mental illnesses I Decreased rates of suicide attempts - Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. II Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006} found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of $1.3-3.1 million per suicide prevented (Lang, 2013). 17 Create Protective Environments Etatinnale] Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddon, 1980}. Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2015; et al., 2016), therefore, changes to the organizational culture in these occupations, by way of implementation of supportive policies or even physical modi?cations to the workplace environment, can demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide, are not (K. L. Knox et al., 2010: National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, changing the availability of and access to common suicide risk factors in the environment such as bridges, alcohol, medications, or ?rearms, can reduce suicide rates, particularly in times of crisis {Beautrais, Gibb, Fergusson, Horwood, 84 Larkin, 2009; Crosby, Espitia~Hardeman, Ortega, 8.: Lozano, 2013; Kaplan et al., 2013: Miller, Warren, Hemenway, 8t Azrael, 2015; C. W. Runyan et al., 2015]. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. 4- Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanginglsuffocation, or jumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1) the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer at al., 2009; Simon et al., 2001} and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007,- Yip et al., 2012). Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches redUcing access to lethal means for persons at?risk of suicide: Interval/ling otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures leg, bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include 18 Comment IAI: Not sure if the additions i made to the rationale are too duplicative of info below. If so, go ahead and delete. erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition-?and keeping medicines in a locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts lRowhani-Rahbar, Simonetti, Rivara, 2016; C. W. Runyan et al., 2015i - Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (National Action Alliance for Suicide Prevention Workplace Task Force, 2015]. - Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8: Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {Cherpitel, Borges, 8L Wilcox, 2004). Potential Outcomes - Increase in safe storage of means - Reduction in suicide attempts 0 Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2015), as described below. 19 Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal ofthe barrier, sadly, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016]. Another program, The Emergency Department Counseiing on Access to Lethoi Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et. al {2016) found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara St Martin, 2012). Police suicides were tracked over 12 years and 20 compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara En Martin, 2012i Another example of this approach is the United States Air Force Suicide Prevention Program AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8a Caine, 2003}. Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community?based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewaid, Remer, 2009). 21 Promote Connectedness to Protect Against Suicide Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002}, for instance between individuals peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one's community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community,?l neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all ofwhich helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. I Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 22 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes - Reduction in maladaptive coping attitudes and behaviors 0 Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed I Increase help-seeking behaviors I Positive perception ofadult support Eviden ce Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (6 metropolitan, 12 rural], Wyman et al. (2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism [Branas et al., 2011). 23 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1985), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters (Pollock Williams, 2004}. Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8; Bunney, 2002) and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, 8: Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 24 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 81 Hunter, 2010]. Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program {Wasserman et al., 2014). In a cluster- randomized controlled trial of YAM conducted across 10 European Union countries and 163 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=D.025] at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% (Wasserman et al., 2014). Signs ofSuicide {505) is another school-based prevention program for students aged 13:11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 8: James, 2016}. In a randomized controlled trial, 505 was shown to reduce self-reported suicide attempts at 3-months post intervention among 25 participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression andlor suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling et al., 2016). Finally, the Good Behavior Game (636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the GBG is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2003). In an outcome evaluation of the 656, first graders assigned to GBG reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (386 effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between (336 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for (536 to be delivered with precision, consistency, and teacher support. 656 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years NY) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide) (Herman et al., 2011). The program includes 9- 20 sessions offered in communitynbased settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, lamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8; Hammond, 26 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionallv, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 27 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims ofviolence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults (and youth). I Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t 'slip through the cracks?. These approaches typically employ screening for depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotiine, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 28 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at-risk individuals in treatment I Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained {Substance Abuse and Mental Health Services Administration, 2014]. However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena 8i. Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. II Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid (MHFA) program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders}, possible riskfactors, and where and how to get evidence-based effective help (Kitchener Jorm, 2004). In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener Jorm, 2006). Gatekeeper training has also been a core part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the 29 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GLS trainings had significantly lower youth suicide rates the year following the training implementation (-1.07, p=.03) (Walrath, Garraza, Reid, Goldston, 8: McKeon, 2015) This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 2007 and 2010. Counties implementing GL3 program activities also had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation of the GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% CI, 1.8-8.0 fewer attempts per 1000 youths]; .003; (Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015)}. More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GLS program. Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Core program was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006]. An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8a. Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 8: Ahmedani, 2015). Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also 30 found that participants experienced significant decreases in suicidality over the course of the telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). In another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Applied Suicide intervention Training (ASST), a widelyr implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safety connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful bv the end of their call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 31 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide (Pitman, Osborn, King, 8: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer E: Sonneck, 1998; Niederkrotenthaler 81 Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%_25% reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas 8: Kutcher, 2011}. 32 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andfor suicide contagion. Potential Outcomes I Reduction in mental health-related sequelae IHCFEESE connectedness - Improved coping skills 0 Improved messaging following suicide I Reduction in re-attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow-up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at-risk of suicide. There are a number of treatments with evidence of impact on risk and protective factors for suicide. One example is the improving Moad? Promoting Access to Collaborative Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. It facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase} by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up (Hunkeler et al., 2006; Unutzer et al., 2006) relative to patients who received care as usual. Another example is Collaborative Assessment and Management af5uicidality (CAMS), which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 33 patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (lobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. {Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two?year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006]. ABFT is a program for adolescents aged 12?18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety [Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care {-4.37 vs. -2.34; .001; Additionally, a higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 46.2%; .006; (Diamond et al., 2010). Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months (at 1, 2, 4, 11 weeks and 4, 6, 12, 18 months}. Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from suicide relative to a treatment- 34 as-usual group versus respectively; chi2 13.83, 0.001) (Fleischmann et al., 2008). Another example of treatment to prevent re-attempts involves active foiiow?up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years). These approaches have been found in a meta-analysis conducted by Inagaki et al. (2015) to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8c Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial found of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). Postvention programs are implemented with the goal of providing support to survivors of suicide to reduce their own risk of suicide. One example of a postvention program, StandBy Response Service (StandBy), provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Coma ns, 35 Scuffha m, 2014). In a study by Visser et al. (2014}, StondBv clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services] Cerel Campbell, 2008). Safe messaging following a suicide. Safe messaging after a suicide can help assure that reporting of the event is done in such a way to reduce risk to consumers of news media and other messaging who may be particularly vulnerable. One way to ensure safe messaging following a suicide is to encourage that reporters adhere to media guidelines for reporting on suicides. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guideiines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact ofthe guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually (95% confidence interval: -149 to -13; -2.32, df 54, 0.024] in the Viennese subway system (Niederkrotenthaler Sonneck, 200?} 36 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in creating safe and protective environments where individuals who are at high risk of suicide can easily access the mental healthcare and services they need. 37 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System?All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 38 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, more timely surveillance data, and critical mention in the President?s FY17 budget, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking), and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination?-in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re-attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer 39 levels of the social ecologv, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 40 References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 41 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2016a). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel,1., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8; Marshall, D. (2014). The Continuum of ?Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lace-threatening behavior, 44(6), 591-600. Chapman, A. L., S: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-235. Classen, T. 1., 8: Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi:10.1002lhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524?1526. Coffey, C. E., Coffey, M. 1., 3a Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M, Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28{11), 963072. doi:10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M., Spilsbury, 1. C., 8: Korbin, 1. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negi, 11111142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 81'. Loza no, B. (2013). Alcohol and suicide. Aicohoi; Science, Poiicy and Pubiic Health, 190. Crosby, A. E., Han, 0., Ortega, L., Parks, 5. E., 84 Gfroerer,1. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 1:13 Years United States, 2003*2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directeo' Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Cunningham, P. 1. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff {Miilwood}, 28(3), w490-501. Curtin, S. C., Warner, M., 8L Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., Avendano, M. {2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am 1 Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 42 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on lv?iolence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Arnl 152(2), 311-318. doi:10.1176lappi.ajp.162.2.311 Diamond, (3. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. [2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987'). Suicide aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8L G. Simpson, trans). Glencoe, it: Free work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 3i. \lella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adolesc Health, 45(4), 351?359. Escobedo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accia' Anal Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, 6. A., 8: lskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., \lagi, K. J., Barnes, J., El. Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311?316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263-280. doi:10.1172l1077559506289524 Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, (3., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06}, 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Godoy Garraza, L., Walrath, C., Goldston, D. B., Reid, H., 81. McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. 43 Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M, 82 Bunney, W. E. (2002). Reducing suicide: A nationai imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 333-352. doi:10.1521fsuli2007373338 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Heaith Reports, 95(5), 411-421. Han, B., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 81 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176-186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205-228. Harris, E. C., Barraclough, B. (1998). Excess mortality of mental disorder. Br,f 173, 11-53. Harris, K. M, Carpenter, C., 8: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 3L Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self?poisoning. British Joarnai of 198(4), 309?316. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi101021702215511028.514 Heise, L., 81 Garcia?Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 87'121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 8: WebsteraStratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co?Occurring Internalizing Q, 26(3), 189-201. Hong, J. S., Kral, M. J., Espelage, D. L., Allen?Mea res, P. (2012). The social ecology of adolescentainitiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431-454. 011~0273dy Houle, J. N., 81 Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073?1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ, 332(7536), 259-263. Inagaki, M., Kawashima, T., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. JAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Isometsa, E. T. (2001). autopsy studies--a review. Eur 16(7), 379-385. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi:10.11117j.1943? 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8i. Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124-2133. 44 Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, 5. 6., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. G., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515-2523. Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust 21 40(1), 6-8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi:10.1027l0227? 5910/a000136 Klevens, J., Simon, T. H., at Chen, J. (2012]. Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 27(10), 1937?2002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 1375. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Knox, M. 5., Burkhart, K., 81 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 23u83. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow?up ofdialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. S., QuispeuAgnoli, M., Ouyang, L., Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am Public Health, 101(6), 1139-1146. doi:10.2105fajph.2010.300010 Maimon, D., Browning, C. R., 81 Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. doi:10.117710022146510377878 Mark, T. L., Shern, D. L., Bagalman, J. E., at Cao, Z. (2007}. Banking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, 0. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group 17 States, 2012. Morb Mortal Wkly Rep, 65(25), 641-645. doi:10.15585lm mwr.mm6525a1 Miller, M., Warren, M., Hemenway, D., Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. 45 Mishara, B. L., 8: Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8L Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., 81. Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc SciMed, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Biaeprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: US. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., 8: Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust 21 41. Olfson, M., Gerhard, T., Huang, (2., 5., 81 Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172?1181. Owens, D. (2002). Fatal and non?fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Pena, J. 3., 8L Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 35(6), 614-637. doi:10.1521isuli2006366614 Perron, 5., Burrows, 5., Fournier, M, Perron, P. A., Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health, 103(7), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Violence Abuse, 13(2), 112-132. Pirkis, J., Too, L. S., Spittal, M. J., Krvsinska, K., Robinson, J., 8i. Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994? 1001. Pitman, A., Osborn, D., King, M., 31 Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 36-94. Pollock, L. R., Williams, J. M. (2004). Problem?solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, 0., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. 0., Ireland, M., Borowskv, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdalesc Health, 35(5), 424 e421? 410. Rowhani-Rahba r, A., Simonetti, J. A., 8-1 Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. doi:10.1093fepirevimxv006 46 Runyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., 31. Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West} Emerg Med, 1 8-14. Runyan, 0., Wattam, C., Ikeda, R., Hassan, F., 8; Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano World Report on Violence and Health (pp. 59-86}. Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. 0., 3t Gan non-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review of Sociology, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 8: James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Shepard, D. 5., Gurewich, 0., Lwin, A. K., Reed, G. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., St Maris, R. W. (1995}. The prevention of suicidal behaviors: An overview. Suicide and life- threatening behavior, 25(1), 10-21. Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide the Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., 81. Day, S. X. (2002}. Universal family-focused interventions in alcohol-use disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 223. Stack, 5., 8L Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 32(1), 103?112. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., St Carpenter, M. {2008}. Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136?142. Szumilas, M, at Kutcher, 5. (2011). Post~suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, 0., Hazell, P., . . .Van Heeringen, K. (2001}. The efficacy of problem?solving treatments after deliberate self-harm: meta?analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979?988. Tyrer, P., Reed, G. M., 8.: Crawford, M. J. (2015}. Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717?726. US. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. US. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006}. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10}, 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014}. Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journalof Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015}. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105lAJ PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. 47 Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. Webster-Stratton, C., Hammond, M. [1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 65(1), 93-109. Webster-Stratton, C., Reid, M. 8: Hammond, M. {2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283-302. Webster-Stratton, C. H., Reid, M. Beauchaine, T. (2011). Combining parent and child training for voung children with ADHD. Journal of Clinical Child 8: Adolescent 40l2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible vears teacher and child training programs in high-risk schools. Journal of child and 49{5), 471-483. Widome, R., Sieving, R. E., Harpin, S. A, Hearst, M. 0. {2008). Measuring neighborhood connection and the association with violence in young adolescents. lAdolesc Health, 43(5), 482-489. Wilcox, H. C., Kellam, S. 6., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Anthony, J. C. [2008). The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560??3. Wilson, W. J. {2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A, Brown, C. H., LoMurrav, M., Schmeelk?Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 16534661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, Wu, K., 81 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 48 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Approach/Program, Practice or Policy Suicide Suicide Other Lead Sectors1 Attempts or Riskarotective ldeation Factors for Suicide Strengthen household financial security Strengthen Unemployment bene?t programs 1/ Government economic (local, state, supports Federal) Housing stabilization policies The National Neighborhood Stabilization 1/ Government Program (local, state, Federal) Strengthen Coverage of mental health conditions in health insurance policies access to mental health Mental Health Parityr Laws v/ v? Government care (state, Federal) Health care Reducing access to lethal means among persons at?risk intervening at hot spots Government (local, state, Establish 5 :eiira: protective afe storage practices 1/ u: ea environments Organizational policies and culture Togetherfor Life 1? Businesleabor US Air Force Suicide Prevention Program Best Available Ev' - Government (local, state, Federal) Businesstabor Community?based policies to reduce excessive alcohol use against suicide Alcohol outiet density v? Government (local, state, Federal) Peer norm approaches Promote connectedness Sources 0f Strength Public Health to protect Social Services Community-engagement activities Greening vacant urban spaces Public Health Social emotional learning Youth Aware of Mental Health Program Public Health Signs of Suicide Education Good Behavior Game solving skills Parenting skill and family relationship approaches The incredible Years Public Health Education Strengthening 10-14 Best Available Evidence identify and Gatekeeper training Mental Health First Aid Public Health Healthcare Social Services Screening combined with care management support people Henry Ford Perfect Depression Core (Pre- Health ca re at-risk cursor to Zero Suicide) Crisis Intervention Notionoi Suicide Prevention Lifeiine v? Public Health Social Services Appiieo' Suicide intervention Training Treatment for people at risk of suicide improving Mood Promoting Access to in? Healthcare Coiioborotive Treatment Social Services Coiioborotive Assessment and Management of Suicidoiity CA M3) -st Available Evidence Interue no to Dialectical Behaviorai Therapy 1/ v" lessen harms and prevent Attachment-Based Famiiy Therapy 1/ future risk Treatment to prevent re-attempts ED Brief interven tion with Follow-up Visits 1/ Health ca re Active folio w?up con tact approaches 1/ 1/ CBTfor Suicide Prevention Postvention 5 tandBy Response Service 1" Health ca re Safe messaging following a suicide Media Guideiines Public Health 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. Policies, Programs, and Practices to Support Individuals, Families, 8: Communities: A Technical Package to Prevent Suicide Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 Poiicies, Programs, and Practices to Support Individuals, Communities: A Technicai Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercv, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, EL, Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technical Package to Prevent Suicide. Atlanta, GA: National Center for Injurv Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Strengthen Access to Mental Health Care 16 Create Protective Environments 18 Promote Connectedness 22 Teach Coping and Problem-Solving Skills 24 Identity and Support People At-Risk 28 Intervene to Lessen Harms and Prevent Future Risk 32 Sector Involvement 37 Monitoring and Evaluation 39 Conclusion 40 References 42 Appendix A 50 Acknowledgments [to be inserted later] External Reviewers [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business/labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8: Melanson, 2011). Self-directed violence may be suicidal or non-soicid'ail in nature. For the purposes of this document, we refer only to behavior where suicide is intended: 0 Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. it contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8: Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, et al., 2011). For example, during 2014, among adults aged 13 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicide (Le. ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model-- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, violence victimization, genetic and biological determinants, hopelessness Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, familyfloved one?s history of suicide, financial and work stress Community level: Inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). lt is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (U.S. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8: Villaveces, 2004; Klevens, Simon, Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, lrwin, Spilsbury, Korbin, 2007; Freisthler, Merritt, 8; LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8.1 Kim, 2012), intimate partner violence (Pinchevsky Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 8a. Ramiro, 2002), intimate partner violence (Heise 8c Garcia-Moreno, 2002; Pinchevsky 8c Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011}. Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness increases individual?s and communities? resilience to suicide and other forms of violence, including connectedness to one?s community (Basile, Hamburger, Swahn, 8: Choi, 2013; Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 8 2007; Kleiman, Riskind, Schaefer, 8-1 Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, Hearst, 2008), school (Basile, Espelage, Rivers, McMahon, 8t Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Br Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 8: Allen-Meares, 2012; Losel Farrington, 2012), family (Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8i Vella-Zarb, 2009; Maimon, Browning, Brooks-Gunn, 2010; Resnick, Ireland, 8: Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8; Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel Farrington, 2012). The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8a Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). By one conservative estimate, for every death by suicide six people are directly impacted survivors). Based on this figure it is estimated that there are over 13 million survivors in the U.S. and unfortunately, survivorship itself is a risk factor for suicide (Crosby 8i. Sacks, 2002). Research indicates that the health consequences of violence, including suicide, are also much more extensive than injury and death. Suicide attempt survivors those with lived experience) may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt (Chapman Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8: Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, BL Carpenter, 2008). The economic toll ofsuicide is immense as well. The total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8r Silverman, 2016). The overwhelming burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (U.S. Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting (Silverman Maris, 1995; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family?, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor] (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014}. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. 10 Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports a Strengthen household financial security Housing stabilization policies Strengthen access to mental health care Coverage of mental health conditions in health insurance policies Create protective environments - Reducing access to lethal means among persons at-risk ofsuicide - Organizational policies and culture It Community-based policies to reduce excessive alcohol use Promote connectedness a Peer norm approaches . Community engagement activities Social-emotional learning Parenting skill and family relationship approaches Gatekeeper training Screening combined with care management Crisis intervention Intervene to lessen harms and prevent Treatment for people at-risk of suicide risk I Treatment to prevent re-attempts I Postvention I Safe messaging following a suicide Teach coping and problem?solving skills identify and support people at risk The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. 11 Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support are profoundly complex and may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide?related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress, increase an individual's risk for suicide; buffering these risks can therefore, potentially protect against suicide (Stack 8L Wasserman, 2007). For example, strengthening economic support systems can help people pay stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes I Reduced suicide rates I Lower foreclosure rates I Lower eviction rates I Reduced emotional distress 13 Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household financial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits (Cylus, Glymour, 3: Avendano, 2014). Another U.5. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk (Classen Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. In terms of lives saved, Flavin 8c Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased their per capita spending on these types of assistance by $45 per year {Flavin 3i. Radcliff, 2009). Housing stabilization policies. The Nationoi Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults {Houle Light, 2014). Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found 14 that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access to Mental Health Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, Stroup, 2015; Owens, 2002), and risk conferred by mental illnesses differ (Arsenault? Lapierre, Kim, 8: Turecki, 2004; E. C. Harris 81 Barraclough, 1997; Tyrer, Reed, 8: Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 84 Barraclough, 1998; World Health Organization, 2014). Studies suggest that up to 90% of people who die by suicide may have had a mental illness at the time of their deaths {Arsenault?Lapierre et al., 2004; Cavanagh, Carson, Sharpe, 8: Lawrie, 2003; Isometsa, 2001). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8: Cao, 2007). Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions {Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention (World Health Organization, 2014). Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law includes coverage for some mental health conditions but not others). Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes I Increased utilization of mental health services 16 I Decreased of mental illnesses I Decreased rates of suicide attempts - Decreased rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. II Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006} found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year and a cost savings of $1.3-3.1 million per suicide prevented (Lang, 2013). 17 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes (Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide (Dahlberg 8i Krug, 2002; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2016; McIntosh et al., 2016], therefore, changes to the organizational culture in these occupations, by way of implementation of supportive policies or even physical modifications to the workplace environment, can change social norms, encourage help?seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (K. L. Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis [Beautrais, Gibb, Fergusson, Horwood, 8i Larkin, 2009; Crosby, Espitia- Hardeman, Ortega, 8i. Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i. Azrael, 2015; C. W. Runyan et al., 2016}. Approaches The current evidence suggests three promising approaches for creating environments that protect against suicide. - Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury]. Research also indicates that 1) the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001) and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between the thought and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening ot Suicide HotSpots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges and cliffs), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include 18 erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition-?and keeping medicines in a locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts lRowhani-Rahbar, Simonetti, Rivara, 2016; C. W. Runyan et al., 2015i - Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (National Action Alliance for Suicide Prevention Workplace Task Force, 2015]. - Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8: Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {Cherpitel, Borges, 8L Wilcox, 2004). Potential Outcomes - Increase in safe storage of means - Reduction in suicide attempts 0 Reduction in suicide deaths I Increase in help-seeking I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness reducing access to lethal means for person at-risk of suicide and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016). 19 Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal ofthe barrier, sadly, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseiing on Access to Lethoi Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre?post quality improvement project, Runyan et al (2016} found that at post-test (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 8c Martin, 2012). Police suicides were tracked over 12 years and 20 compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.42 suicides per 100,000 population per year compared to an 11% increase in the control city (rate: 29.0 per 100,000} (Mishara En Martin, 2012i Another example of this approach is the United States Air Force Suicide Prevention Program AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8a Caine, 2003}. Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community?based policies to reduce excessive alcohol use. While multiple policies to limit alcohol use exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewaid, Remer, 2009). 21 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951]. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002}, for instance between individuals peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social refers to a sense of trust in one's community and neighborhood, social integration, and also the availability and participation in social organizations {Beyen Layde, Hamberger, 3r. Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013}. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured by social trust, community/i neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all ofwhich helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole {Centers for Disease Controi and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. - Peer norm approaches seek to normalize prosocial behaviors/protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 22 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes - Reduction in maladaptive coping attitudes and behaviors 0 Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed I Increase help-seeking behaviors I Positive perception ofadult support Eviden ce Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (6 metropolitan, 12 rural], Wyman et al. (2010} found that the program improved peer leaders' adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (Wyman et al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant associated reductions in community residents' self-reported stress levels and engagement in more physical exercise than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism [Branas et al., 2011). 23 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories {Bandura, 1985), surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness] characteristics. The literature linking life skills and suicide is The inability to employ adequate coping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attempters (Pollock Williams, 2004}. Treatments that include bolstering problem skills (Goldsmith, Pellmar, Kleinman, 8; Bunney, 2002) and include problem-solving techniques (Ghahramanlou-Holloway, Bhar, Brown, Olsen, 8: Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 81 Webster-Stratton, 2011). - Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 24 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 81 Hunter, 2010]. Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide 0 Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program {Wasserman et al., 2014). In a cluster- randomized controlled trial of YAM conducted across 10 European Union countries and 163 schools, students participating in the YAM program were significantly less likely to have an incident suicide attempt 0.45, 0.24?0.85; p=0.014} and severe suicidal ideation (0.50, 0.27-0.92; p=D.025] at the 12-month follow-up compared to the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and RR fell by 49.6% (Wasserman et al., 2014). Signs ofSuicide {505) is another school-based prevention program for students aged 13:11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 8: James, 2016}. In a randomized controlled trial, 505 was shown to reduce self-reported suicide attempts at 3-months post intervention among 25 participating students compared to control students. The SOS program also increased students? knowledge of how to get help for themselves or friends for depression andlor suicidal thoughts, and favorable attitudes toward help-seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to lower?risk participants (Schilling et al., 2016). Finally, the Good Behavior Game (636) is a classroom-based program for elementary school children aged 6-10; it represents an example of upstream suicide prevention programming. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the GBG is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2003). In an outcome evaluation of the 656, first graders assigned to GBG reported half the adjusted odds of suicidal ideation and suicide attempts. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (386 effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of (336 students, neither suicidal ideation nor suicide attempts were significantly different between (336 and the control interventions (Wilcox et al., 2008). This finding likely arose due to the lack of implementation fidelity and pointed to the need for (536 to be delivered with precision, consistency, and teacher support. 656 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years NY) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide) (Herman et al., 2011). The program includes 9- 20 sessions offered in communitynbased settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, lamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8; Hammond, 26 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionallv, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening has been shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 27 Identify and Support People At-Risk Rationale In order to decrease suicide, attention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously, individuals who are institutionalized, have been victims ofviolence, or are homeless; and members of certain ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. a Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults (and youth). I Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t 'slip through the cracks?. These approaches typically employ screening for depression andfor suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotiine, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 28 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at-risk individuals in treatment I Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained {Substance Abuse and Mental Health Services Administration, 2014]. However, there is limited evidence for effectiveness screening programs, but at the same time, standard principles for public health screening make them promising (Pena 8i. Caine, 2006). The number of studies evaluating crisis intervention services is limited, but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. II Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid (MHFA) program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises and/or in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders}, possible riskfactors, and where and how to get evidence-based effective help (Kitchener Jorm, 2004). In a randomized controlled trial of 300 participants of MHFA, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener Jorm, 2006). Gatekeeper training has also been a core part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the 29 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GLS trainings had significantly lower youth suicide rates the year following the training implementation (-1.07, p=.03) (Walrath, Garraza, Reid, Goldston, 8: McKeon, 2015) This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 2007 and 2010. Counties implementing GL3 program activities also had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation of the GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% CI, 1.8-8.0 fewer attempts per 1000 youths]; .003; (Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015)}. More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GLS program. Screening combined with care management and overall continuity of care. The Henry Ford Perfect Depression Core program was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006]. An examination of the impact of the Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention (1999 and 2000) to the intervention years (2002-2009). During this time period, the suicide rate fell 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8a. Ahmedani, 2013). Further, suicide rates also declined among HMO members who participated in targeted suicide prevention efforts and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased (M. Coffey, Coffey, 8: Ahmedani, 2015). Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline (NSPL) to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also 30 found that participants experienced significant decreases in suicidality over the course of the telephone session, and that levels of hopelessness and pain continued to decrease after their initial call (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007). In another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Applied Suicide intervention Training (ASST), a widelyr implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safety connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful bv the end of their call to the hotline. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 31 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and/or who have made suicide attempts andfor have engaged in non-suicidal self-injury are at increased risk of suicide Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide (Pitman, Osborn, King, 8: Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion (Etzersdorfer E: Sonneck, 1998; Niederkrotenthaler 81 Sonneck, 2007). Approaches A broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. 0 Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%_25% reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015) I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas 8: Kutcher, 2011}. 32 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide andfor suicide contagion. Potential Outcomes I Reduction in mental health-related sequelae IHCFEESE connectedness - Improved coping skills 0 Improved messaging following suicide I Reduction in re-attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow-up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at-risk of suicide. There are a number of treatments with evidence of impact on risk and protective factors for suicide. One example is the improving Moad? Promoting Access to Collaborative Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. It facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase} by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up (Hunkeler et al., 2006; Unutzer et al., 2006) relative to patients who received care as usual. Another example is Collaborative Assessment and Management af5uicidality (CAMS), which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 33 patient about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (lobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. {Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two?year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006]. ABFT is a program for adolescents aged 12?18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety [Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care {-4.37 vs. -2.34; .001; Additionally, a higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 46.2%; .006; (Diamond et al., 2010). Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 18-months (at 1, 2, 4, 11 weeks and 4, 6, 12, 18 months}. Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five culturally different sites found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from suicide relative to a treatment- 34 as-usual group versus respectively; chi2 13.83, 0.001) (Fleischmann et al., 2008). Another example of treatment to prevent re-attempts involves active foiiow?up contact approaches such as postcards, letters, and telephone calls, are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post-discharge with some programs continuing contact for 2 or more years). These approaches have been found in a meta-analysis conducted by Inagaki et al. (2015) to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8c Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. it uses a risk reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial found of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). Postvention programs are implemented with the goal of providing support to survivors of suicide to reduce their own risk of suicide. One example of a postvention program, StandBy Response Service (StandBy), provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Coma ns, 35 Scuffha m, 2014). In a study by Visser et al. (2014}, StondBv clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively, 0.005). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services] Cerel Campbell, 2008). Safe messaging following a suicide. Safe messaging after a suicide can help assure that reporting of the event is done in such a way to reduce risk to consumers of news media and other messaging who may be particularly vulnerable. One way to ensure safe messaging following a suicide is to encourage that reporters adhere to media guidelines for reporting on suicides. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guideiines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact ofthe guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually (95% confidence interval: -149 to -13; -2.32, df 54, 0.024] in the Viennese subway system (Niederkrotenthaler Sonneck, 200?} 36 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, health care insurers and providers, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Access to Mental Health Care. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 37 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People At?Risk and to Lessen Harms and Prevent Future Risk. The intensity and activities of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health care, social services, and justice sectors can work collaboratively to support individuals at high- risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 38 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System?All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 39 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of prevention programs on preventing suicide, as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And while suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis--at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to namejust a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear ofasking someone about their risk of suicide [versus the fear and consequence of not asking}, and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burden. Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination-?in a multi-level, multi-sectoral way. This technical package includes strategies and approaches targeting upstream prevention, social emotional learning for children and youth, as well as strategies focused more cognitive behavioral treatment to prevent re-attempts. It includes universal, selective, and indicated strategies, or strategies that focus on the whole population regardless of risk to strategies that focus on those groups at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches at the outer 4O levels of the social ecologv, policies to stabilize housing and community engagement initiatives. In short, care and attention has been paid to all aspects of suicide prevention. While the evidence base continues to be built, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible.? 41 References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 42 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2016a). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel,1., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8; Marshall, D. (2014). The Continuum of ?Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lace-threatening behavior, 44(6), 591-600. Chapman, A. L., S: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-235. Classen, T. 1., 8: Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi:10.1002lhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524?1526. Coffey, C. E., Coffey, M. 1., 3a Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M, Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28{11), 963072. doi:10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M., Spilsbury, 1. C., 8: Korbin, 1. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negi, 11111142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 81'. Loza no, B. (2013). Alcohol and suicide. Aicohoi; Science, Poiicy and Pubiic Health, 190. Crosby, A. E., Han, 0., Ortega, L., Parks, 5. E., 84 Gfroerer,1. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 1:13 Years United States, 2003*2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directeo' Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Cunningham, P. 1. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff {Miilwood}, 28(3), w490-501. Curtin, S. C., Warner, M., 8L Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., Avendano, M. {2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am 1 Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 43 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on lv?iolence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Arnl 152(2), 311-318. doi:10.1176lappi.ajp.162.2.311 Diamond, (3. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. [2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987'). Suicide aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8L G. Simpson, trans). Glencoe, it: Free work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 3i. \lella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adolesc Health, 45(4), 351?359. Escobedo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accia' Anal Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, 6. A., 8: lskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., \lagi, K. J., Barnes, J., El. Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311?316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263-280. doi:10.1172l1077559506289524 Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, (3., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06}, 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Godoy Garraza, L., Walrath, C., Goldston, D. B., Reid, H., 81. McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. 44 Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M, 82 Bunney, W. E. (2002). Reducing suicide: A nationai imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 333-352. doi:10.1521fsuli2007373338 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Heaith Reports, 95(5), 411-421. Han, B., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 81 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176-186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205-228. Harris, E. C., Barraclough, B. (1998). Excess mortality of mental disorder. Br,f 173, 11-53. Harris, K. M, Carpenter, C., 8: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 3L Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self?poisoning. British Joarnai of 198(4), 309?316. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi101021702215511028.514 Heise, L., 81 Garcia?Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 87'121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 8: WebsteraStratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co?Occurring Internalizing Q, 26(3), 189-201. Hong, J. S., Kral, M. J., Espelage, D. L., Allen?Mea res, P. (2012). The social ecology of adolescentainitiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431-454. 011~0273dy Houle, J. N., 81 Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073?1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ, 332(7536), 259-263. Inagaki, M., Kawashima, T., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. JAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Isometsa, E. T. (2001). autopsy studies--a review. Eur 16(7), 379-385. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi:10.11117j.1943? 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8i. Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124-2133. 45 Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, 5. 6., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. G., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515-2523. Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust 21 40(1), 6-8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi:10.1027l0227? 5910/a000136 Klevens, J., Simon, T. H., at Chen, J. (2012]. Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 27(10), 1937?2002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 1375. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Knox, M. 5., Burkhart, K., 81 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 23u83. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow?up ofdialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. S., QuispeuAgnoli, M., Ouyang, L., Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am Public Health, 101(6), 1139-1146. doi:10.2105fajph.2010.300010 Maimon, D., Browning, C. R., 81 Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. doi:10.117710022146510377878 Mark, T. L., Shern, D. L., Bagalman, J. E., at Cao, Z. (2007}. Banking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, 0. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group 17 States, 2012. Morb Mortal Wkly Rep, 65(25), 641-645. doi:10.15585lm mwr.mm6525a1 Miller, M., Warren, M., Hemenway, D., Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. 46 Mishara, B. L., 8: Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8L Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., 81. Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc SciMed, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Biaeprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: US. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., 8: Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust 21 41. Olfson, M., Gerhard, T., Huang, (2., 5., 81 Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172?1181. Owens, D. (2002). Fatal and non?fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Pena, J. 3., 8L Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 35(6), 614-637. doi:10.1521isuli2006366614 Perron, 5., Burrows, 5., Fournier, M, Perron, P. A., Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health, 103(7), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Violence Abuse, 13(2), 112-132. Pirkis, J., Too, L. S., Spittal, M. J., Krvsinska, K., Robinson, J., 8i. Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994? 1001. Pitman, A., Osborn, D., King, M., 31 Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 36-94. Pollock, L. R., Williams, J. M. (2004). Problem?solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, 0., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. 0., Ireland, M., Borowskv, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdalesc Health, 35(5), 424 e421? 410. Rowhani-Rahba r, A., Simonetti, J. A., 8-1 Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. doi:10.1093fepirevimxv006 47 Runyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., 31. Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West} Emerg Med, 1 8-14. Runyan, 0., Wattam, C., Ikeda, R., Hassan, F., 8; Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano World Report on Violence and Health (pp. 59-86}. Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. 0., 3t Gan non-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review of Sociology, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 8: James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Shepard, D. 5., Gurewich, 0., Lwin, A. K., Reed, G. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., St Maris, R. W. (1995}. The prevention of suicidal behaviors: An overview. Suicide and life- threatening behavior, 25(1), 10-21. Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide the Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., 81. Day, S. X. (2002}. Universal family-focused interventions in alcohol-use disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 223. Stack, 5., 8L Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 32(1), 103?112. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., St Carpenter, M. {2008}. Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136?142. Szumilas, M, at Kutcher, 5. (2011). Post~suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, 0., Hazell, P., . . .Van Heeringen, K. (2001}. The efficacy of problem?solving treatments after deliberate self-harm: meta?analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979?988. Tyrer, P., Reed, G. M., 8.: Crawford, M. J. (2015}. Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717?726. US. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. US. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006}. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10}, 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014}. Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journalof Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015}. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105lAJ PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. 48 Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. Webster-Stratton, C., Hammond, M. [1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 65(1), 93-109. Webster-Stratton, C., Reid, M. 8: Hammond, M. {2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283-302. Webster-Stratton, C. H., Reid, M. Beauchaine, T. (2011). Combining parent and child training for voung children with ADHD. Journal of Clinical Child 8: Adolescent 40l2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible vears teacher and child training programs in high-risk schools. Journal of child and 49{5), 471-483. Widome, R., Sieving, R. E., Harpin, S. A, Hearst, M. 0. {2008). Measuring neighborhood connection and the association with violence in young adolescents. lAdolesc Health, 43(5), 482-489. Wilcox, H. C., Kellam, S. 6., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Anthony, J. C. [2008). The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560??3. Wilson, W. J. {2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A, Brown, C. H., LoMurrav, M., Schmeelk?Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 16534661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, Wu, K., 81 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 49 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Other Risk] Protective Factors for Suicide Suicide Attempts or ldeation Lead Sectors1 Government (local, Unempioym ent benefit programs state, Federal) Strengthen economic Other income supports Businessflabor supports . . . . . . Housmg pollcles Government {locaL The National Neighborhood state, Federal) Program Coverage of mental health conditions in health insurance policies Health ca re Strengthen Mental Heaith Parity Laws 1/ access to Government (state, mental health Fadera? ca re Reducing access to lethal means among persons at-risk Government (local, intervening at hot spots state) 1/ Safe storage practices Public Health Establish protective Organlzatlonal pollcles and culture Busmesstabor . Together for Life enwronments Government (local, US Air Force Suicide Prevention Program 1/ state, Fed eral) 50 Best Available Evidence Community?based policies to reduce excessive alcohol use Government {local, state) Promote connectedness to protect against suicide Alcohol outlet density 1/ Businessllabor Peer norm approaches Public Health Sources of Strength Education Communitvvengagement activities Public Health Greening vacant urban spaces Government {local} Mental Health First Aid Social emotional learning Public Health Youth Aware of Mental Health Program Education Signs of Suicide Good Behavior Game 50"",ng skills Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening Families 10-14 Gatekeeper training Pu blic Health Healthcare 51 identify and support people at-risk Screening combined with care management Best Available Evidence Henry Ford Perfect Depression Care {Pre- cursor to Zero Suicide) Healthcare Social Services Crisis Intervention Public Health National Suicide Prevention Lifeline Social Services Applied Suicide in tervention Skills Training v? Treatment for people at risk of suicide Healthcare improving Mood Promoting Access to Social Services Collaborative Treatment Justice Collaborative Assessment and Management 1/ Saicldalit CA MS Intervene to ?l lessen harms Dialectical Behavioral Therapy and prevent future ?5k Attachment?Based Family Therapy Treatment to prevent re-attempts Health ca re ED Brief intervention with Follow?up Visits Social Services Active folio w-up contact approaches CBTfor Suicide Prevention 52 Best Available Evidence Postvention Health ca re StondBy Response Service Safe messaging following a suicide Public Health Media Guidelines 9? Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. Title A Technical Package to Prepared by: Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 [Title] is a publication of the National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technical Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents External Reviewers [to be inserted later] Acknowledgments [to be inserted later] 'r Comment IAI: Changed the word I ?harms? here to effects since we use harms in the next sentence. [Comment IAI: New sentences I: {Comment New sentence i I Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to i prevent suicide. Broadlv, the strategies represented include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long?term Eaffects ]of suicidal behaviorfor individuals, families, communities, and societv. Specificallv, the strategies include environments; promoting connectedness to protect against suicide; teaching coping and problem- solving skills; identifying and supporting people at?risk; and intervening to lessen harms and prevent future risk. [This package supports the National Strategv for Suicide prevention, Goal 1, "Integrate and coordinate suicide prevention activities across multiple sectors and settings." (p.29) it also supports the National Action Alliance for Suicide Prevention?s priority "To create and disseminate a framework for communitv~based suicide prevention." comprehensive implementation ofthis package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome (Frieden, 2014}. Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategv. This can be accomplished through programs, policies, and: practices. [The approaches included come primarilv from studies based in the United States. __hee_vid_en_cej for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined bv the CDC, is part of a broader class of behavior called self-directed vioi'ence. Self- directed violence refers to behavior directed at oneself that deliberately.I results in injurv or the potential for injury Self-directed violence mav be suicidai? or non-suicidai in nature. For the purposes of this document, we refer onlv to behavior where suicide is intended. Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotof self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicidal behavior presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (2, In 2014 in the U.S., the most recent death data available, suicide was responsible for 42,773 deaths, which is approximately a suicide every 12 minutes (4). In 2014, suicide ranked as the tenth leading cause of death and has been among the top twelve leading causes of death since 1925 in the U5. (5) Overall suicide rates have increased from 1999 to 2014 {24% increase) Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (4). Suicides reflect only a portion of the number of persons affected by suicidal thoughts and behaviors Substantially more persons are hospitalized as a result of nonfatal suicidal behaviors than are fatally injured, and an even greater number are either treated in ambulatory settings or not treated at all For example, during 2014, among adults aged 213 years, for every one suicide there were 9 adults treated in hospital emergency departments for self-inflicted injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide Suicides, attempts, and ideation take an immense emotional, physical, and economic toll on individuals, families and communities. By one estimate, for every death by suicide six people are directly impacted (Le. survivors}. Based on this figure it is estimated that there are over 13 million survivors in the US. and unfortunately, survivorship itself is a risk factor for suicide (9). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. It occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another and act cumulatively to increase one?s vulnerability to think about or engage in suicidal behaviors. The social-ecological model is a useful framework for viewing and understanding suicidal risk factors that have been identified in the literaturel: Relationship Individual Risk Factors for Suicide;3 Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, previous victimization, acute and chronic stressors financial problems), genetic and biological determinants, hopelessness Relationship: High conflict or violent relationships, sense of isolation and lack of social support, family history of suicide, financial and work stress Community: Inadequate community connectedness, barriers to health care-- lack of access to providers or medications Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking An individual having or experiencing one or a number of risk factors does not always lead to suicide; for example, the vast majority of individuals who are depressed do not die by suicide. It is also important to note that these risk factors described above is not an exhaustive list. These and many other risk factors exist and can be arranged differently or contribute to multiple areas within the social-ecological model. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status?. Protective factors for suicide: Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model and may vary in significance depending on the context in which the suicide occurs. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, peers, and family, connectedness to school, community and other social institutions and the availability of physical and mental health care?. These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide is connected to other forms of violence in a number of different ways. First, suicide and other forms of violence often share some of the same root causes. They can all take place under one roof, or in a given community or neighborhood and can happen at the same time or at different stages of life?? For example, in neighborhoods where there is low cohesion, or where residents don't support and trust each other, people are at higher risk for suicideiil as well as perpetration of child teen dating violence,? intimate partner violence,?iL and youth Lack of economic opportunities and unemployment are associated with suicide, as well as perpetration of child maltreatment,Xi intimate partner sexual violencelivand youth violence.W Also, while most people who are victims of violence do not act violently or die by suicide, people who experience or are exposed to suicide are at a higher risk for both being a victim of other forms ofviolence and for inflicting harm on others. For example, children who experience physical abuse or neglect early in their lives are at greater risk for suicide,?i and also at greater risk for committing violence against peers {particularly for teen dating violence,m and committing child abuse,? elder intimate partner and sexual violencexi?l? later in life. There are also a number of protective factors that pose an opportunity to protect individuals and communities from suicide and other forms of violence, and buffer the effects ofshared risk factors. For example, connectedness increases peoples? and communities' resilience to suicide and other forms of violence, including connectedness t0 ?Li" ca ring ad pro-social The health and economic consequences of suicide are substantial. Total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (10). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exist and are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting. Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family-, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor}. According to CDC Director, Tom Frieden, successful public health programs also require political commitment, funding, communication, and performance monitoring. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta?analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on Specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Eontext and Cross?Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such 10 that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports Strengthen financial security Housing stabilization policies Strengthen access to mental health care Coverage of mental health conditions in health insurance policies Establish protective environments Means restriction Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness to protect against suicide Peer norm approaches Community engagement activities Teach coping and problem-solving skills Social-emotional learning Parenting skill and family relationship approaches identify and support people at risk Gatekeeper training Screening combined with care management Crisis intervention intervene to lessen harms and prevent future risk Treatment for people at-risk of suicide Treatment to prevent reda?empts Postvention Safe messaging following a suicide The example programs, policies. and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. [Add text for other issues to potentially cover in this section. You may want to take a look at the other packages in this regard] 11 This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business/labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Approaches Brief statement introducing the approaches {see example statements in the other technical packages). Strengthen household financial security. Research indicates that economic crises are related to suicide rates. Findings from the U5. show that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Policies that support financial security during difficult economic times have been shown to mitigate the risk of economic crises on suicide rates. Housing stabilization policies that aim to strengthen housing stability and security may help to buffer the impact of foreclosures and evictions on suicide, as recent research has drawn an association between housing instability and suicidal behavior. Programs that provide affordable housing and other options for homebuyers such as loan modification programs may be used in conjunction with move-out planning and financial counseling services to minimize the impact of foreclosures and evictions on suicide. 13 Potential Outcomes Evidence Brief statement introducing the evidence [see example statements in the other technical packages]. I Strengthen household financial security. I Housing stabilization policies. 14 Strengthen Access to Mental Health Care Rationale Mental illness is a risk factor for suicide. Studies suggest that up to 90% of people who die by suicide may have had a mental illness [Cavanaugh et al., 2003). Research on state-level suicide rates have been found to be correlated with general mental health measures such as depression (Arsenault-Lapierre, et al., 2004). While most people with mental health problems do not attempt or die by suicide, assuring access to quality mental health care is critical to suicide prevention. Approaches A major approach to strengthening access to mental health care is to have health insurance policies that include coverage for such services. 0 Coverage of mental health conditions in health insurance policies. Health insurance policies that allow people with mental health problems to access mental health treatment in the same way that they access health care for physical health concerns can increase use of mental health services, help normalize treatment seeking in the population, reduce of mental illnesses like depression and bipolar disorder, and in turn, reduce rates of suicide and suicide attempts. Potential Outcomes I Increased access to mental health services I Decreased of mental illnesses Decreased rates of suicide attempts I Decreased rates of suicide Evidence Brief statement introducing the evidence. I Coverage of mental health conditions in health insurance policies. Using data from the National Survey of Drug Use and Health, Harris et al. (2005} found that 12 months after states enacted parity laws, self-reported use of mental healthca re services significantly increased. Subsequent research by Lang et al. (2011), suggests that mental health parity laws are associated with an approximate 5% reduction in suicide rates. This reduction (in 29 states} equated to the prevention of 592 suicides per year and a cost savings of $133.1 million per suicide prevented {Lang, 2013). 15 Establish Protective Environments Rationale Suicide prevention efforts that focus on both the individual and hisy?her environment increase the likelihood of lives saved. Establishing protective environments helps ensure that all of the places where an individual lives, works, and plays are supportive. Limiting access to lethal means, be it at home or in nature, and particularly when an individual may be most vulnerable, can literally make the difference between life and death. Likewise, creating a work environment conducive to prevention and focused on employee well?being supports the large majority of the population where they spend much of their day. Finally, policies that reduce the availability of alcohol, a potent suicide risk factor, serve to support individuals and protect the environment further. Approaches The current evidence suggests three approaches with promise for creating environments that protect against suicide. These include: 0 Means Restriction. Modifying the environment to decrease access to lethal means is an important public health strategy for preventing suicide. Acute suicidal crises are often brief and impulsive. Previous research indicates that the interval between thinking about suicide and attempting can be as short as 5-10 minutes [Simon et al., 2001,; Deisenhammer et al., 2009}. Getting past the impulse by making it more difficult to access lethal means can be lifesaving. Highly lethal means such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and have high fatality rates about 85% of people who use a firearm in an attempt die from their injury}. Research also indicates that most people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007; Yip et al., 2012}. Removing or restricting access to lethal means changes the context of the potential suicide and whether the outcome will be fatal or non-fatal (Yip et al., 2012} intervening at Suicide Hotspots. These interventions are focused on preventing suicides at locations which offer direct means for suicide or a secluded place that prevents intervention. Suicide hotspots include tall structures (for example, bridges and cliffs}, railway tracks, and isolated locations that are popular destinations for suicide {for example, parks}. Interventions include barriers to prevent jumping and signs and telephones to encourage suicidal individuals to seek help. 0 Sofa Storage Practices for medications, firearms, and other household products can reduce the risk for suicide by preventing impulsive action and separating individuals from easy access to lethal means. Safe storage practices include education and counseling around storing firearms locked in a secure place in a gun safe or lock box}, preferably unloaded and separate from the ammunition. Keeping medicines in a 16 locked cabinet or secure location can also prevent their misuse by children and adolescents. I Organizational policies and culture that focus on prosocial behavior, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and that have leadership support from the top down can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation). I Community-based policies to reduce excessive alcohol use. Acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges 8n Wilcox, 2004}. While various community policies exist to reduce excessive alcohol use zoning limits related to alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age}, previous research indicates that policies related to outlet locations and densitv are more strongly associated with suicide, making these particular policies an important approach to preventing suicide. Potential Outcomes I Increase in safe storage of lethal means I Reduction in suicide attempts I Reduction in suicide deaths I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016). I Means restriction intervening or suicide hotspots A meta-analysis of suicide hotspot interventions implemented in combination or in isolation in the US. and abroad found they reduced suicide (Pirkis et al., 2015; Cox, 2013). For example, suicide deaths from jumping from the Jacques?Cartier bridge in Canada decreased after the installation of a bridge barrier {incidence rate ratio 0.24; 95% confidence interval 0.13, 0.43], which persisted when all bridges 0.39; 95% CI 02?, 0.55} and all jumping sites 0.66; 95% CI 0.54, 0.80) in the regions were considered, suggesting little or no displacement to other jumping sites (Perron, et. al., 2013]. in contrast, the 17 removal of safety barriers on the Grafton Bridge in Auckland, New Zealand led to an immediate and substantial increase in both the numbers and rate of suicide byjumping from the bridge from 3 to 15 (p 0.01); the rate of such deaths also increased ip<0.01; Beautrais, 2001; Beautrais, et. al., 2009}. 0 Safe storage practices In a case-control study of firearm-related events identified by medical examiner and coroner offices from 37 counties in Washington, Oregon, and Missouri, and 5 trauma centers, Grossman et al., [2005) found that safe storage practices storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device) were protective for suicide attempts among adolescents (Grossman et al., 2005}. A recent systematic review of clinic and community?based education and counseling around safe storage of firearms found that the provision of safety devices significantly increases safe firearm storage practices compared to counseling alone or providing economic incentives to acquire safety devices {Rowhani-Rahbar, Simonetti, Rivara, 2016). The Emergency Department Counseling on Access to Lethal Means (ED CALM) program trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post study, Runyan et al (2015) found that among the parents contacted at follow-up, 76% reported all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at the time of the child?s initial emergency department visit, all reported guns were currently locked, compared to 57% reporting this at the time of the initial visit. Organizational policies and culture Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components of Togetherfar Life were designed to foster an organizational culture that promoted mutual support and solidarity among members of the Force, help for problems related to suicide, training of supervisors, managers and all units to improve competencies in identifying suicidal risk and in using existing resources, and an education campaign to improve awareness and help?seeking. After implementation of the program, police suicides were tracked over 12 years and compared to rates in the control city of Quebec. Pre- post assessments of learning, interviews, and focus groups were also included. The suicide rate in the intervention group decreased significantly by 78.9% [p .008} to 6.42 per 100,000 per year compared to 29.0 per 100,000 in the control citya?a significant difference in rates {p (Mishara 8 Martin, 20121 The United States Air Force Suicide Prevention Program inclusive of 11 policy and education initiatives, was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training}, and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service-wide problems impacting the whole community. Using a time-series design, the AFSPP program 18 was associated with a 33% relative risk reduction in suicide [Knox et al., 2003). The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch] found significantly lower rates of suicide after the program was launched than before [Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years {Knox et al., 2010). Community-based policies to reduce excessive alcohol use Several studies on alcohol outlet density in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Giesbrecht et al., 2014; Escobedo 8i. Ortiz, 2002). For example, Giesbrecht et al. found that both on and off?premises alcohol outlets restaurants where alcohol is served and stores where alcohol is available for purchase to go) were positively associated with alcohol?related suicides in 14 U.S. states, particularly among men 1.08, and American Indiaanlaska Natives (ADE: 1.36; CI: Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 yea rs in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald 3t Remer, 2009). These findings suggest that measures to reduce alcohol outlet density can potentially reduce alcohol~involved suicides. 19 Promote Connected ness to Protect Against Suicide Rationale The ouantitv and qualitv of our social connection with others has been linked with suicide dating as far back to Durkheim, who first posited that the weakening social bonds is among the chief causes for suicidalitv?. Connectedness is the degree to which an individual or group of individuals are sociallv close, interrelated, or share resources with othersz. Connectedness can be formed within and between multiple levels of the social ecologv3; for instance between individuals leg. peers, neighbors, co? workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Connectedness is also viewed as a broad term that encompasses several other concepts including social support, social participation, social isolation, social cohesion, social capital, social integration, all of which have been linked to suicidal behaviors either conceptuallv or scientificallv within the literature. lconnectedness can serve either to protect or mitigate suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversitv. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primarv prevention activities to the community as a wholez. Insert info on connectedness and ind?udesome or all at the following re social capital here].50cial capital refers to a sense of trust in one?s communitv and neighborhood, social integration, and also the availabilityr and participation in social organizations (Muenning et al., 2013; Bever et al., 2014]. Manv ecological cross?sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generallv, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured bv social trust, communitv/neighborhood engagement, and improved mental healtl] Approaches The following two approaches are designed to promote connectedness among individuals and within communities to protect against suicide. Peer norm approaches seek to normalize help-seeking, encourage reaching out and talking to trusted adults, and promote supporting peers through building connectedness. These approaches are tvpicallv delivered in school settings but can also be implemented in community settingsComment IAI: This is what was in the table that Linda said could be added to the rationale. Community engagement activities. Community engagement is an aspect of social capital and involves residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes Reduction in maladaptive coping attitudes and behaviors Increase in healthy coping attitudes and behaviors Increase in referrals for youth in distressed Increase help-seeking behaviors Positive perception of adult support Evidence Brief statement to introduce the evidence. Current evidence provide some support for these types of approaches for reducing risk factors associated with suicidal behaviors. I Peer norm approaches. Evaluations show that programs such as Sources of Strength (505) can improve school norms and beliefs about suicide that are created and disseminated by student peers. in a randomized controlled trial of the program conducted with 18 urban and rural high- schoois Wyman et al., (2010fl found that $05 improved peer leaders' knowledge of adult supports for suicidal peers and were more likely than untrained leaders to refer distressed peers to adult supports Trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders I Community-engagement activities. Greening vacant urban spaces is an initiative that communities have adopted, particularly after the Great Recession (200?~2009) when many cities across the U.S. experienced urban abandonment. These initiatives engaged community members in the cleaning, greening, or beautifying vacant areas. One vacant lot greening initiative, which greened 4,436 lots in 4 areas of Philadelphia, PA, greening significantly reduced community residents' self-reported stress levels ip<.001} and engaged in more physical exercise ip<.001) than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism? 21 Teach Coping and Problem-Solving Skills Rationale Building life skills prepare individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem- solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors?. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theoriesf, surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness} characteristics. The literature linking life skills and suicide is The inability to employ adequate ceping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attemptersa. Treatments that include bolstering problem skills9 and include problem?solving techniquesmv11 appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: - Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work. I Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse. 22 Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide I Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) I Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills 0 Enhance problem-solving and conflict management skills Evidence There are a several programs with evidence that supports teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. These programs, commonly delivered in schools, overall have demonstrated success in preventing suicide risk. However, the evidence is mixed; while most models demonstrated strong effects, others were mixed. For example, Signs of Suicide is a high school-based prevention program for students designed to increase knowledge about suicide improve and normalize help-seeking behaviors. After 3 months of participation, 505 has documented a 54% reduction in suicide attempts among its participants compared to controls. 505 participants with a lifetime history of suicide attempt were also significantly less likely to report planning a suicide in the 3 months following the program compared to lower-risk participants. Moreover, increased knowledge of how access help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help-seeking were noted among students who participated in 505?. Another example is The Good Behavior Game {636), which is a classroom-based program for elementary school children uses a behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. In the evaluation of this program, several outcome behaviors were measured including suicide, given its association with externalizing behavior problems. Results from the first cohort evaluation found that first graders assigned to 686 reported half the adjusted odds of suicide ideation and suicide attempts than controls. After 14 years postvention, 636 was also found to significantly reduce risk of substance abuse?. However, in a replication trial, 636 did not produce the same reductions on suicide ideation or attempts as it did in the first cohort?. This finding likely arose due to the lack of implementation fidelity and pointed to the need for (336 to be delivered with precision, consistency, and teacher support 23 Finally, the Youth Aware of Mental Health Program (YAM), a school-based program rigorously evaluated in 10 European countries, teaches youth about suicide risk and protective factors and problem-solving skills. At 12 months, students participating in YAM were significantly less likely to have a suicide attempt and severe suicidal ideation compared to controls; Risk of suicide ideation and attempts fell by 49 to 54% among YAM participants?. Parenting skill and family relationship programs Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. One program, The incredible Years is a universally delivered program for parents of children up to 12 years of age designed to improve responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence and reduce problem behaviors in children. Several studies have demonstrated the effect of The incredible Years program on reducing internalizing such as anxiety and depression, and child conduct The program is also associated with improved problem-solving and conflict management, skills that were maintained at 1-yearfollow-up17'13'19. The program also demonstrated greater benefits as the dosage of the intervention increased?. Strengthening Families 10-14 years is a program that involves sessions between parents, youth, and family with the goals of: 1) improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; 2) promoting youths? interpersonal and problem-solving skills; and 3) creating family activities to build cohesion and positive parent-child interactions. The program has successfully shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating familiesu. 24 Identify and Support People At-Risk Rationale In order to be successful in decreasing suicidal behavior, attention must be paid to those who are at-risk or vulnerable. These persons experience risk and occurrence of suicidal behavior that is higher than average. This group requires particular focus on proactive case finding and retention and access to services. These vulnerable or disadvantaged populations include (but are not limited to): those living in lower socio-economic status; members of certain ethnic minority groups; those with a mental health problem; those who are institutionalized; those who have been victims of violence; and those who are homeless. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care are still key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need ofthem. Nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches This document outlines three approaches that focus on identifying and supporting those who are at?risk- - Gatekeeper training is typically implemented in schools and within health care settings and is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care to assure that people who may be at high-risk of suicide don?t 'slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow-up. Programs such as these have demonstrated beneficial effects on depression, suicide ideation, and suicide mortality. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means restriction, 25 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts 0 Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at?risk individuals in treatment 0 Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained. {Substance Abuse and Mental Health Services Administration, Center for Mental Health Services. Report to Congress: Garret Lee Smith Suicide Prevention Program. US Department of Health and Human Services: Rockville, MD, 2014). There is limited evidence for effectiveness screening programs but standard principles for public health screening make them promising. (Pena JB, Caine ED. Screening as an Approach for Adolescent Suicide Prevention. Suicide and Life-Threatening Behavior. 2006; The number of studies evaluating crisis intervention services is limited but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. Gatekeeper training. Mental Health First Aid (MHFA), designed for the lay public, consists of three weekly sessions of three hours each. The content covers helping people in mental health crises and/or in the early stages of mental health problems. The crisis situations covered included suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior. The mental health problems discussed included depressive, anxiety and disorders. The co-morbidity with substance use disorders is also covered. Participants learn the of these disorders, possible risk factors, where and how to get help and evidence-based effective help. In a randomized controlled trial of 300 participants, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. An additional finding was an improvement in the mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 26 2004). Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener 81 Jorm, 2006). Gorret Lee Smith (GL5) Suicide Prevention Program, as of June 2014, 154 GL5 grants had been awarded to 49 states and 48 tribes. Gatekeeper training has been a core part of all GL5 programs, and grantees have consistently reported spending the largest proportion of their budget on this strategy. A multi-site evaluation assessed the connection between community gatekeeper training (activities such as the number of people trained and the intended mid- and long-term outcomes) and a reduction of suicide attempts and deaths. To address this question, the analysis compared the change in the suicide mortality rates and nonfatal suicidal behavior amongthe population aged youth and young adults in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Effect on fatal behavior When compared with similar counties that did not implement GLS training, counties implementing GLS trainings presented significantly lower youth suicide rates the year following the training implementation (-1.07, This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 200? and 2010. Effect on nonfatal behavior - Counties implementing GL5 program activities had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation ofthe GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% Cl, 1.8-8.0 fewer attempts per 1000 youths]; .003). More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GL5 program. Walrath C, Godoy-Garraza L, Reid H, Goldston DB, McKeon R. impact of the Garrett Lee Smith Youth Suicide Prevention Program on Suicide Mortality. American Journal of Public Health: May 2015, Vol. 105, No.5, pp. 986-993. doi: Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 2015; 72:1143-1149. Brief statement introducing the evidence. 1- Screening combined with care management and overall continuity of care. Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) The overall goal of the program was to eliminate suicide. More broadly, the aim of the program was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety. The redesign focused on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each ofthese aims. The program began with screening and assessment of each patient for suicide 27 risk with coordinated continuous follow-up care system wide. On the basis of the combined total of 31 suicides for the 11-year observation period, the rate of suicide among patients was 97 per 100,000 lN=13l for the two baseline years {the average rate for 1999 and 2000). This rate is similar to that reported for a clinical population. For the start-up year (2001), the rate of suicide was 41 per 100,000 For the follow?up interval {the average for 2002?2009], the rate was 19 per 100,000 Poisson regression analysis showed a statistically significant decrease of 82% in the suicide death rate between the baseline (1999?2000) and intervention (2002?2009) years (rate 95% confidence 535.001}. Coffey et al., 2013; Coffey, 2006. Additionally, between 1999 and 2010, researchers found that suicide rates declined among HMO members who received mental health specialty services, in association with a target prevention effort, and increased among HMO members who accessed general medical services but not specialty MH services (Coffey et. al., 2015]. Crisis intervention. Notionoi Suicide Prevention Lifeline (NSPL). This is a nationwide hotline that operates 24/? and is accessible by phone or a web-based chat function. Trained counselors are on-hand to listen, offer fee and confidential emotional support, and provide referrals for mental health services in the local area. Suicide callers lN=1,085) were assessed during their calls to the hotline and 380 participated in the follow-up assessment. Results indicated that seriously suicidal individuals called the hotline. Significant decreases in suicidality were found during the course of the telephone session, with continuing decreases in hopelessness and pain in the following weeks (Gould et al., 2007). Applied Suicide intervention Training This a training program for hotline counselors, emergency workers, clergy, caregivers and others in the community. The ASIST model has three phases of caregiving: connecting, understanding and assisting. The training helps participants identify people who are having thoughts of suicide and to recognize their invitation for help (connecting); to listen to the caller?s reasons for dying and living (understanding); and how to conduct a safety assessment, develop a safety plan for the person at risk, and connect the person at risk to community resources (assisting). The ASIST training program has been field tested in a variety of settings. in a national randomized controlled trial, Gould et al (2013) assessed the impact of the ASIST training across the NSPL network of hotlines over the period 2008-2009. Data were derived from 1,507 monitored calls from 1,410 suicidal individuals to 17 Lifeline centers. Callers who spoke with ASIST?trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of calls compared to those who spoke to counselors receiving usual care training. ASIST-trained counselors were also better able to keep callers on the phone longer and 28 establish a connection. ASIST training, however, did not yield more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 29 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and have had non-fatal suicide attempts or have engaged in non-suicidal self-injury are at increased risk of subsequent suicide- related morbidity and mortality. Risk of suicidality can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide. Exposure to sensationalized or uninformed reporting regarding suicide-related deaths may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Approaches A broad array of approaches to lesson harms and reduce future risk of suicidality among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. I Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. I Treatment to prevent re?attempts. These follow-up contact approaches use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors to prevent reattempts. These approaches typically focus on coping and other emotional regulation skills and may include case management home visits to increase adherence to and continuity of care, one?on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%-25% reattem pt within a year, and of attempt survivors die by suicide within 1 to 5 years oftheir initial attempt (lnagaki, et al., 2015}. I Postvention approaches are implemented after a suicide has taken place and may include debriefing of survivors [those who have lost a friend, peer, family member, co-worker to suicide}, counseling, and/'or bereavement support groups. The programs have not typically tested their impact on suicide or suicidal behavior but may reduce risk of guilt, feelings of depression, and complicated grief {Szumilas et al., 2011). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (no outreach) {Cerel 8: Campbell, 2003i 30 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Potential Outcomes I Reduction in mental health-related sequelae Increase I Improved coping skills 0 Improved messaging following suicide 0 Reduction in re-attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk ofsuicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation of suicide-related mortality is a statistically rare event, evaluation of mortality outcomes requires large sample sizes and extended follow-up. Therefore, much of the evidence in this area primarily focuses on rislt and protective factors. Treatment for people at-risk of suicide. has been demonstrated to help diverse individuals in different settings to reduce risks related to mental health problems such as depression and anxiety, and to reduce suicidal ideation and attempts. Evaluation evidence for selected approaches to prevent future suicide risk follow below. Improving Mood Promoting Access to Collaborative Treatment (IMPACT) aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. IMPACT facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase) by a depression care manager. IMPACT has been shown to significantly improve quality of life, and to reduce functional impairment, depression (Hunkeler et al., 2006; Uniitzer et al., 2002) and suicidal ideation over 24smonths of follow-up (Uniitzer et al., 2006} relative to patients who received care as usual. 31 Collaborative Assessment and Management of Suicidality (CAMS) is a therapeutic framework for suicide-specific assessment and treatment of patient?s suicide risk. it is a flexible approach that can be used across treatment settings and clinician theoretical orientations. The clinician and patient work together in an interactive assessment process. The patient is highly engaged in the development of their own treatment plan. Every session of CAMS is collaborative and involves the patient?s input about what is and is not working. Ultimately, this process is designed to enhance the therapeutic alliance and increase treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies (lobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community- based sample of suicidal outpatients. (Comtois et al., 2011). Dialectical Behavioral Therapy is a multicomponent therapy for individuals at high risk for suicide who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self-injurious behavior, those receiving DBT were half as likely to make a suicide attempt at twonyear follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self?injurious acts combined (Linehan et al., 2006). Attachment-Based Family Therapy (ABFT) is a program for adolescents ages 12-18 that is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety. A randomized controlled trial of ABFT (Diamond et al., 2010} found that suicidal adolescents assigned to ABFT experienced significantly greater improvement regarding suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care (-4.37 vs. -2.34; .001; d=0.97l. A higher percentage of ABFT participants reported no suicidal ideation in the past week at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 46.2%; .006; Treatment to prevent re?attempts. Active contact and follow-up approaches intended to prevent reattempts among patients that have been hospitalized and subsequently discharged for suicide attempts have been found in a meta-analysis conducted by lnagaki et al., {2015) to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect 32 of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of postcrisis suicide prevention long-term follow-up contact approach, Motto et al., [2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of followup than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam et al., 2011; Wang et al., 2016]. I Postvention approaches such as StondBy Response Service provide suicide bereavement support services to clients via face~to~face outreach and telephone support delivered by a professional crisis response team. A site coordinator develops a customized case management plan, referring clients to other existing community services matched to their needs. In a study by Visser et al. (2014], StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and Edi/E. respectively, 0.005). 0 Safe messaging following a suicide. Media guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrUpted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually (95% confidence interval: -149 to ?2.32, df 54, 0.024] in the Viennese subway system [Niederkrotenthalen T., Sonneck, (3., 2007) Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population?level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non?governmental organizations. Collectively, these sectors can 33 make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and in?uential role 34 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law Enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair et al., 2016). The National Electronic Injury Surveillance System-All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.}, age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 35 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion 36 References 37 Appendix A: Summary of Strategies and Approaches to Prevent uicid St rategv Approach} Program, Practice or Policy Strengthen ?nancial security Best Available Evidence Suicide Suicide Other Attempts or Risk! Protective Ideation Factors for Suicide Lead Sectors? 1 Comment help you complete the lead sector column. For the other columns, you just need to insert a check? mark based on the evidence vou describe In the narrative for a particular program or policy. For example, if the evidence shows impact on suicide, then put a check-mark in that column. If the studvr also found effects on risk or protective factors, then put a check-mark in that Strengthen Unemployment benefit programs economic supports Housing stabilization policies The National Neighborhood Program Strengthen Coverage of mental health conditions in health insurance policies BCCESS to mental health care Mentoi Heol'th Parity Lows Establish protective environments Means restriction intervening at hot spots Safe storage practices it (attempts) it [Safe storage of ?rearms and med ication) Organizational policies and culture Tag ether for Life US Air Fa rce Suicide Prevention Pro gram [familv violence) Lcolumn as well. Best Available Evidence Communitywbasecl policies to reduce excessive alcohol use Aicohoi outiet density Peer norm approaches Promote connectedness Sources of Strength Ni to protect against suicide Communitvrengagement activities Greening vacant urban spaces ?i Social emotional learning Youth Aware of Mentai Heaith Program xi Signs of Suicide *1 Teach co in 3 Good Behavior Game Ni and problem- Parenting skill and family relationship approaches The incredibie Years xi Strengthening 10-14 Gatekeeper training Mentai Health First Aid Best Available Evidence Screening combined with care management Identify and support people Henry Ford Perfect Depression Care {Pre- at-risk cursor to Zero Suicide) Crisis Intervention Nationai Suicide Prevention Lifeiine Appiied Suicide intervention Training Treatment for people at risk of suicide improving Mood - Promoting Access to Caiiabarative Treatment Coiiaharative Assessment and Management Intervene to ofSuicidaiity (CAMS) lessen harms and prevent future risk Diaiecticai Behaviorai Therapy Attachment-Based Famiiy Therapy Treatment to prevent re-attempts ED Brief intervention with Foiiow-up Visits Active foiiaw?up con tact approaches CBTfor Suicide Prevention Best Available Evidence Postvention StondBy Response Service Safe messaging following a suicide Media uideiines - 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. i Butchart A, Phinney A, Check P, Villaveces A. Preventing violence: a guide to implementing the recommendations of the World Report on Vioience ond Heoith. Geneva, Switzerland: World Health Organizaiton: 2004. Klevens J, Simon TR, Chen J. Are the perpetrators of violence one and the same? Exploring the co?occurrence of perpetration of physical aggression in the United States. i interpers Vioience. Desai, R. A, Dausey, D. J.. Rosenheck, R. A. (2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. American Journal of i" Coulton Cl, Crampton D5, Irwin M, Spilsbury 1C, Korbin JE. How neighborhoods influence child maltreatment: a review of the literature and alternative pathWays. Child Abuse Negl. Freisthler B, Merritt DH, LaScaIa EA. Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat. Capaldi DM, Knobic NB, Shortt JW, Kiln A systematic review of risk factors for Intimate partner violence. PartnerAbuse. Pinchevskv GM, Wright EM. The impact of neighborhoods on intimate partnerviolence and victimization. Trauma Violence Abuse. Sampson RJ, MorenoffJD, Gannon-Rowiey T. Assessing "neighborhood effects": Social processes and new directions in research. Annual review of sociology. 2002;23:443473. ?1 Luo F, Florence CS, Quispe-Agnoli M, Duvang L, Crosby AE. Impact of business cycles on US suicide rates, 1928-2001 American journal of public health. Reeves A, Stuckler D, McKee M, Gunnell D, Chang SS, Basu S. Increase in state suicide rates in the USA during economic recession. The Lancet. 1814. "l Runyan D, Wattam, C, Ikeda, R., Hassan, F, Ramiro, L. Child abuse and neglect by parents and caregivers. In: Krug E, Dahlberg LL, Mercy, JA, Zwi AB, Lozano R, ed. World report on violence and health. Geneva, Switzerland: World Health Organization; 200259-86. Heise L, GarciarMoreno C. Violence by intimate partners. In: In: Krug E, Dahlberg LL, Mercy, JA, Zwi AB, Lozano R, ed. World report on vioience ond heoith. Geneva, Switzerland: World Health Organization; 200231121 GM, Wright EM. The impact of neighborhoods on intimate partner violence and victimization. Trauma Violence Abuse. Centers for Disease Control and Prevention, Sexual Violence: Risk and Protective Factors. Available at: Accessed April 4, 2014. Wilson When work disappears: The world of the new urban poor. New York, NY: Random House 2011. ?i Briere, ., Madni, L. A., 82: Godbout, N. (2015). Recent Suicidality in the General Population Multivariate Association With Childhood Maltreatment and Adult Victimization. Journal ofinierpersonai vioience, 0886260515584339. Logan E, Lech RT, Barker LE. Gender-specific mental and behavioral outcomes among physically abused high-risk seventh-grade youths. PaiJiic Reports. Duke NN, Pettingell 5L, McMorris Bl, Borowskv IW. Adolescent violence perpetration: associations with multiple types of adverse childhood experiences. Pediatrics. Duke NN, Pettingell SL, McMorris BJ, Borowskv IW. Adolescent violence perpetration: associations with multiple types of adverse childhood experiences. Pediatrics. Violence and tire Famiiy: Report of'ihe. American Association Presidentiai Task Force on Vioience and the Family. Washington, DC: American Association; l996. Vioience and the Famiiv.? Report ofihe American Association Presidentiai Task Force on Vioience and the arniiv. Washington, DC: American Association; 1996. Vioience and the Famiiy: Report of'rne American Association Presiden-iiair Task Force on Vioience and tire Famiiy. Washington, DC: American Association; 1996. ewkes R. Rape perpetration: a review. Pretoria. Serbia! Violence Research initiative. 2012. ?it Couiton, C. J., Crampton, D. 8.. Irwin, M., Spilsbury, J. C., Korbin. J. E. How neighborhoods influence child maltreatment: A review of the literature and alternative pathways. Child abuse 8. negiect, 31(11), 1117-1142. Pinchevsky GM, Wright EM. The impact of neighborhoods on intimate partner violence and victimization. Trauma Vioience Abuse. 12?132. m? Basile KC, Hamburger ME, Swahn MH, Choi C. Sexual violence perpetration by adolescents in dating versus same?sex peer relationships: differences in associated risk and protective factors. The Western Journal of Emergency Medicine. m? Widome R, Sieving RE, Harpin SA, Hearst MD. Measuring neighborhood connection and the association with violence in young adolescents. Journal of Adolescent Health. 2008;43l5lz482-489. mi? Kieiman, E. M, Riskind. J. H., Schaefer, K. E., Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide l'lSk. Ci?iSiS. ?it Centers for Disease Control and Prevention, Elder Maltreatment: Risk and Protective Factors. Available at: Accessed April 4, 2014. Borowskv IW, Hogan M, Ireland M. Adolescent sexual aggression: risk and protective factors. Pediatrics. mi Capaldi DM, Knoble NB, Shortt JW, Kim A systematic review of risk factors for Intimate partner violence. Partner Abuse. m? DeGue S, Massetti GM, Holt MK, et Identifying links between sexual violence and youth violence perpetration. ofVioience. mm Basile KC, Espelage DL, Rivers 1, McMahon PM, Simon TR. The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and Miriam Behavior. ?xi" Liiscl F, Farrington DP. Direct protective and buffering protective factors in the development ot?youth violence. American Joarnai of Preventive Medicine. ZDI Hong JS, Espelage DL. A review ofresearch on bullying and peer victimization in school: An ecological system analysis. Aggression and Vioienr Behavior. Carter, M., McGee, R., Taylor, 8.. 8: Williams, S. {2007). Health outcomes in adolescence: Associations with family. friends and school engagement. Journal of adolescence, 30(1), 51-62. Centers for Disease Control and Prevention. Child Maltreatment: Risk and Protective Factors. Available at: Accessed April 4., 2014. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A systematic review of risk factors for Intimate partner violence. Partner Abuse. ml" Resnick. M. D., Ireland, Ivi., Borowsky, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. Journal of adolescent health, 35(5), 424-e1. Elgar Fl, Craig W, Boyce W, Morgan A, Vella?Zarb R. Income inequality and school bullying: multilevel study of adolescents in 37 countries. Journal of Adolescent Health. ll" Maimon, D., Browning, C. R., 8; Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Journal ofbealth and soclal behavlor, 51(3), 307-324. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A systematic review of risk factors for Intimate partner violence. Partner Abuse. 1?280. Losel F. Farrington DP. Direct protective and buffering protective factors in the development of youth violence. American Journal of Preventive Medicine. Maimon, D.. Browning. C. R., Brooks-Gum, J. {2010). Collective ef?cacy, family attachment. and urban adolescent suicide attempts. Journal of health and soclal behavior, 51(3), BOT-324. ll? Capaldi DM, Knobel NB, Shortt JW, Kim HK. A systematic review of risk factors for intimate partner violence. Partner Abuse. Losel F, Farrington DP. Direct protective and buffering protective factors in the development of youth violence. American Journal of Preventive Medicine. l. Crosby AE, Ortega L, Melanson C. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. 2. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press. 2014. 3. US. Surgeon General, National Action Alliance for Suicide Prevention. 2012 National strategy for suicide prevention: Goals and Objectives for Action. 2012.Nat Strategy 4. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web-based Injury Statistics Query and Reporting System [onlinej. (2005) Accessed on 20161une 3. Available from URL: 5. Centers for Disease Control and Prevention. Ten leading causes of death in the United States, 1975. Atlanta: Public Health Service; 1978. 6. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?20142. NCHS data brief, no 241. Hyattsville, MD: National Center for Health Statistics. 2016. 7. Crosby AE, Han B, Ortega LAG, Parks SE, Gfroerer J. Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008e2009. MMWR 2011:60(No. 8. Ferdon C, Crosby AE, Caine ED, Hindman J, Reed GA, lskanderJ. CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR In press 9. Crosby AE, Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide 8t lifewthreatening behavior. Fall 10. Shepard DS, Gurewich D, Lwin AK, Reed GA Jr, Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav. 2015; 45:352-62. doi: 10.1111/sltb.12225. Epub 2015 Oct 29. Kann, McManus, Harris et al. Youth Risk Behavior Surveillance United States, 2015. MMWR Surveill Summ 2016;65 (No. References for Shane?s section of risk and protective factors: 1- Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 200221?56 2- World Health Organization. World Health Organization. Geneva, Switzerland: World Health Organization, 2014. 3' U.S. Department of Health and Human Services (HHS) Office of the Surgeon General and National Action Alliance for Suicide Prevention. Strotegyfor Suicide Prevention: Goods and Objectives for Action. Washington, DC: HHS, September 2012. Zalsman G. Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10?year systematic review. Lancet Ref: Blair JM. Fowler KA, Jack SPD, Crosby A5 The National Violent Death Reporting System: overview and future directions. in) Prev 2016;22:3uppl 1 i6?i'l 1. NEISS AIP Ref: CDC, Web-based Injury Statistics Query and Reporting System, Nonfatal Injury Reports, 2001-2014. Available at: Shane?s section: 1- Durkheim E. Suicide: A study in sociology. Glencoe Press: New York; 1951. 2- Centers for Disease Control and Prevention (CDC) (2011). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from 10. 11. 12. 13. 14. Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56 Wyman, P. A., Brown, C. H., Loiv?lurray, M, Schmeelk?Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Pubiic Health, 100(9), 1653-1661. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., &Ten Have, T. R. (2011). A difference-in-differences analysis of health, safety, and greening vacant urban space. American Journal of Epidemiology, 174(11), 1296-1306. World Health Organization. World Health Organization. Geneva, Switzerland: World Health Organization, 2014. Bandura A (1986) Social Foundations of Thought and Action: ASocial Cognitive Theory. Englewood Cliffs: NewJersey: Prentice-Hall Publishers. Pollock, LR. and Williams, MG. (2004). Problem-solving in suicide attempters. Medicine, pp 163-162. Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M., 8: Bunney, W. E. (2002). Reducing suicide: A notiondi imperative: National Academies Press. Townsend, E., Hawton, K., Altman, 0., Arensman, E., Gunnell, D., Hazell, P., . . . Van Heeringen, K. (2001). The efficacy of problem- solving treatments after deliberate self?harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. medicine, 31(6), 979-988. M. Ghahramanlou-Holloway, S. S. Bhar, G. K. Brown, C. Olsen and A. T. Beck (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42, pp 1185-1193. Schilling, E. A., Aseltine, R. H., Jr., 81 James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Kellam, S. (3., Brown, C. H., Poduska, J. M, Ialongo, N. S., Wang, W., Toyinbo, P., et al. (2008}. Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug and Alcohol Dependence, 55-528. H.C. Wilcox, S.G. Kellam, C.H. Brown, J. Poduska, N.S. Ialongo, W. Wang, J. Anthony. The impact of two universal randomized first- ancl second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 (2008), pp. 560-- 573 15. 16. 18. 19. 2D. 21. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, V. (2015). School-based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. doi: 10.1016/50140- Webster-Stratton, C., Reid, M. 1., Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: Evaluation of the Incredible Years teacher and child training programs in high-risk schools. Journal of Child and 49(5), 47'1-488. Reid, M. 1., Webster?Stratton, (3., 8a. Hammond, M. (2003}. Follow-up of children who received the Incredible Years intervention for oppositional-de?ant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34, 471-491. Webster-Stratton, C., Hammond, M. (1997). Treating children with early-onset conduct problems: A comparison of child and parenting training interventions. Journal of Consulting and Clinical 65(1), 93-109. Webster-Stratton, C., Reid, M. 1., Hammond, M. (2001). Preventing conduct problems, promoting social competence: A parent and teacher training partnership in Head Start. Journal of Clinical Child 30(3), 283-302. Herman, K. C., Borden, L., Reinke, W. M., S: Webster- Stratton, C. (2011). The impact of the Incredible Years Parent, Child, and Teacher Training Programs on children's co-occurring internalizing School Quarterly, 26, 189-201. doi:10.1037/ a0025228. Spoth, R., Guyll, M., Day, 5. X. (2002). Universal family-focused interventions in alcohol-use disorder prevention: Cost-effectiveness and cost-benefit analyses of two interventions. Journal of Studies on Alcohol, 63(2), 219228. Title A Technical Package to Prepared by: Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 [Title] is a publication of the National Center for Injury Prevention and Control ofthe Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for lnjurvr Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: [authors] A Technical Package to Atlanta, GA: National Center for Injuryr Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents External Reviewers [to be inserted later] Acknowledgments [to be inserted later] 'r Comment IAI: Changed the word I ?harms? here to effects since we use harms in the next sentence. [Comment IAI: New sentences I: {Comment New sentence i I Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to i prevent suicide. Broadlv, the strategies represented include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long?term Eaffects ]of suicidal behaviorfor individuals, families, communities, and societv. Specificallv, the strategies include environments; promoting connectedness to protect against suicide; teaching coping and problem- solving skills; identifying and supporting people at?risk; and intervening to lessen harms and prevent future risk. [This package supports the National Strategv for Suicide prevention, Goal 1, "Integrate and coordinate suicide prevention activities across multiple sectors and settings." (p.29) it also supports the National Action Alliance for Suicide Prevention?s priority "To create and disseminate a framework for communitv~based suicide prevention." comprehensive implementation ofthis package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome (Frieden, 2014}. Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategv. This can be accomplished through programs, policies, and: practices. [The approaches included come primarilv from studies based in the United States. __hee_vid_en_cej for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined bv the CDC, is part of a broader class of behavior called self-directed vioi'ence. Self- directed violence refers to behavior directed at oneself that deliberately.I results in injurv or the potential for injury Self-directed violence mav be suicidai? or non-suicidai in nature. For the purposes of this document, we refer onlv to behavior where suicide is intended. Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotof self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicidal behavior presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (2, In 2014 in the U.S., the most recent death data available, suicide was responsible for 42,773 deaths, which is approximately a suicide every 12 minutes (4). In 2014, suicide ranked as the tenth leading cause of death and has been among the top twelve leading causes of death since 1925 in the U5. (5) Overall suicide rates have increased from 1999 to 2014 {24% increase) Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?19 years, also second among persons in their 205 and 305; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 (4). Suicides reflect only a portion of the number of persons affected by suicidal thoughts and behaviors Substantially more persons are hospitalized as a result of nonfatal suicidal behaviors than are fatally injured, and an even greater number are either treated in ambulatory settings or not treated at all For example, during 2014, among adults aged 213 years, for every one suicide there were 9 adults treated in hospital emergency departments for self-inflicted injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide Suicides, attempts, and ideation take an immense emotional, physical, and economic toll on individuals, families and communities. By one estimate, for every death by suicide six people are directly impacted (Le. survivors}. Based on this figure it is estimated that there are over 13 million survivors in the US. and unfortunately, survivorship itself is a risk factor for suicide (9). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is complex with no single determining cause. It occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another and act cumulatively to increase one?s vulnerability to think about or engage in suicidal behaviors. The social-ecological model is a useful framework for viewing and understanding suicidal risk factors that have been identified in the literaturel: Relationship Individual Risk Factors for Suicide;3 Individual: History of depression and other mental illnesses, alcohol and drug abuse, previous suicide attempt, previous victimization, acute and chronic stressors financial problems), genetic and biological determinants, hopelessness Relationship: High conflict or violent relationships, sense of isolation and lack of social support, family history of suicide, financial and work stress Community: Inadequate community connectedness, barriers to health care-- lack of access to providers or medications Societal: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking An individual having or experiencing one or a number of risk factors does not always lead to suicide; for example, the vast majority of individuals who are depressed do not die by suicide. It is also important to note that these risk factors described above is not an exhaustive list. These and many other risk factors exist and can be arranged differently or contribute to multiple areas within the social-ecological model. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status?. Protective factors for suicide: Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model and may vary in significance depending on the context in which the suicide occurs. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, peers, and family, connectedness to school, community and other social institutions and the availability of physical and mental health care?. These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide is connected to other forms of violence in a number of different ways. First, suicide and other forms of violence often share some of the same root causes. They can all take place under one roof, or in a given community or neighborhood and can happen at the same time or at different stages of life}!2 For example, in neighborhoods where there is low cohesion, or where residents don't support and trust each other, people are at higher risk for suicide3 as well as perpetration of child maltreatment,?5 teen dating violence,Er intimate partner violence,? and youth violence.3 Lack of economic opportunities and unemployment are associated with suicide, 9'1? as well as perpetration of child maltreatment,11 intimate partner violencefl? sexual violence?1 and youth violence.15 Also, while most people who are victims of violence do not act violently or die by suicide, people who experience or are exposed to suicide are at a higher risk for both being a victim of other forms of violence and for inflicting harm on others. For example, children who experience physical abuse or neglect early in their lives are at greater risk for suicide,16 and also at greater risk for committing violence against peers (particularly for boysj,? bullying,? teen dating violence,? and committing child abuse}0 elder abuse,21 intimate partner violence,22 and sexual violence23 later in life. There are also a number of protective factors that pose an opportunity to protect individuals and communities from suicide and other forms of violence, and buffer the effects of shared risk factors. For example, connectedness increases peoples' and communities? resilience to suicide and other forms of violence, including connectedness to one's caring and pro-social peers.45-45 The health and economic consequences of suicide are substantial. Total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (10). Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exist and are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting. Rather, suicide prevention is best achieved by a focus across the individual-, relationship-, family-, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor}. According to CDC Director, Tom Frieden, successful public health programs also require political commitment, funding, communication, and performance monitoring. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta?analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on Specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developmental nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Eontext and Cross?Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such 10 that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports Strengthen financial security Housing stabilization policies Strengthen access to mental health care Coverage of mental health conditions in health insurance policies Establish protective environments Means restriction Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness to protect against suicide Peer norm approaches Community engagement activities Teach coping and problem-solving skills Social-emotional learning Parenting skill and family relationship approaches identify and support people at risk Gatekeeper training Screening combined with care management Crisis intervention intervene to lessen harms and prevent future risk Treatment for people at-risk of suicide Treatment to prevent reda?empts Postvention Safe messaging following a suicide The example programs, policies. and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. [Add text for other issues to potentially cover in this section. You may want to take a look at the other packages in this regard] 11 This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business/labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Approaches Brief statement introducing the approaches {see example statements in the other technical packages). Strengthen household financial security. Research indicates that economic crises are related to suicide rates. Findings from the U5. show that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Policies that support financial security during difficult economic times have been shown to mitigate the risk of economic crises on suicide rates. Housing stabilization policies that aim to strengthen housing stability and security may help to buffer the impact of foreclosures and evictions on suicide, as recent research has drawn an association between housing instability and suicidal behavior. Programs that provide affordable housing and other options for homebuyers such as loan modification programs may be used in conjunction with move-out planning and financial counseling services to minimize the impact of foreclosures and evictions on suicide. 13 Potential Outcomes Evidence Brief statement introducing the evidence [see example statements in the other technical packages]. I Strengthen household financial security. I Housing stabilization policies. 14 Strengthen Access to Mental Health Care Rationale Mental illness is a risk factor for suicide. Studies suggest that up to 90% of people who die by suicide may have had a mental illness [Cavanaugh et al., 2003). Research on state-level suicide rates have been found to be correlated with general mental health measures such as depression (Arsenault-Lapierre, et al., 2004). While most people with mental health problems do not attempt or die by suicide, assuring access to quality mental health care is critical to suicide prevention. Approaches A major approach to strengthening access to mental health care is to have health insurance policies that include coverage for such services. 0 Coverage of mental health conditions in health insurance policies. Health insurance policies that allow people with mental health problems to access mental health treatment in the same way that they access health care for physical health concerns can increase use of mental health services, help normalize treatment seeking in the population, reduce of mental illnesses like depression and bipolar disorder, and in turn, reduce rates of suicide and suicide attempts. Potential Outcomes I Increased access to mental health services I Decreased of mental illnesses Decreased rates of suicide attempts I Decreased rates of suicide Evidence Brief statement introducing the evidence. I Coverage of mental health conditions in health insurance policies. Using data from the National Survey of Drug Use and Health, Harris et al. (2005} found that 12 months after states enacted parity laws, self-reported use of mental healthca re services significantly increased. Subsequent research by Lang et al. (2011), suggests that mental health parity laws are associated with an approximate 5% reduction in suicide rates. This reduction (in 29 states} equated to the prevention of 592 suicides per year and a cost savings of $133.1 million per suicide prevented {Lang, 2013). 15 Establish Protective Environments Rationale Suicide prevention efforts that focus on both the individual and hisy?her environment increase the likelihood of lives saved. Establishing protective environments helps ensure that all of the places where an individual lives, works, and plays are supportive. Limiting access to lethal means, be it at home or in nature, and particularly when an individual may be most vulnerable, can literally make the difference between life and death. Likewise, creating a work environment conducive to prevention and focused on employee well?being supports the large majority of the population where they spend much of their day. Finally, policies that reduce the availability of alcohol, a potent suicide risk factor, serve to support individuals and protect the environment further. Approaches The current evidence suggests three approaches with promise for creating environments that protect against suicide. These include: 0 Means Restriction. Modifying the environment to decrease access to lethal means is an important public health strategy for preventing suicide. Acute suicidal crises are often brief and impulsive. Previous research indicates that the interval between thinking about suicide and attempting can be as short as 5-10 minutes [Simon et al., 2001,; Deisenhammer et al., 2009}. Getting past the impulse by making it more difficult to access lethal means can be lifesaving. Highly lethal means such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and have high fatality rates about 85% of people who use a firearm in an attempt die from their injury}. Research also indicates that most people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007; Yip et al., 2012}. Removing or restricting access to lethal means changes the context of the potential suicide and whether the outcome will be fatal or non-fatal (Yip et al., 2012} intervening at Suicide Hotspots. These interventions are focused on preventing suicides at locations which offer direct means for suicide or a secluded place that prevents intervention. Suicide hotspots include tall structures (for example, bridges and cliffs}, railway tracks, and isolated locations that are popular destinations for suicide {for example, parks}. Interventions include barriers to prevent jumping and signs and telephones to encourage suicidal individuals to seek help. 0 Sofa Storage Practices for medications, firearms, and other household products can reduce the risk for suicide by preventing impulsive action and separating individuals from easy access to lethal means. Safe storage practices include education and counseling around storing firearms locked in a secure place in a gun safe or lock box}, preferably unloaded and separate from the ammunition. Keeping medicines in a 16 locked cabinet or secure location can also prevent their misuse by children and adolescents. I Organizational policies and culture that focus on prosocial behavior, skill building, changing social norms, referral and access to helping services mental health, substance abuse treatment, financial counseling), and that have leadership support from the top down can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation). I Community-based policies to reduce excessive alcohol use. Acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges 8n Wilcox, 2004}. While various community policies exist to reduce excessive alcohol use zoning limits related to alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age}, previous research indicates that policies related to outlet locations and densitv are more strongly associated with suicide, making these particular policies an important approach to preventing suicide. Potential Outcomes I Increase in safe storage of lethal means I Reduction in suicide attempts I Reduction in suicide deaths I Reduction in alcohol-related suicide deaths Evidence The evidence for the effectiveness of means restriction and other ways to establish protective environments is some of the strongest in the field (Zalsman et al., 2016). I Means restriction intervening or suicide hotspots A meta-analysis of suicide hotspot interventions implemented in combination or in isolation in the US. and abroad found they reduced suicide (Pirkis et al., 2015; Cox, 2013). For example, suicide deaths from jumping from the Jacques?Cartier bridge in Canada decreased after the installation of a bridge barrier {incidence rate ratio 0.24; 95% confidence interval 0.13, 0.43], which persisted when all bridges 0.39; 95% CI 02?, 0.55} and all jumping sites 0.66; 95% CI 0.54, 0.80) in the regions were considered, suggesting little or no displacement to other jumping sites (Perron, et. al., 2013]. in contrast, the 17 removal of safety barriers on the Grafton Bridge in Auckland, New Zealand led to an immediate and substantial increase in both the numbers and rate of suicide byjumping from the bridge from 3 to 15 (p 0.01); the rate of such deaths also increased ip<0.01; Beautrais, 2001; Beautrais, et. al., 2009}. 0 Safe storage practices In a case-control study of firearm-related events identified by medical examiner and coroner offices from 37 counties in Washington, Oregon, and Missouri, and 5 trauma centers, Grossman et al., [2005) found that safe storage practices storing firearms unloaded, separate from ammunition, in a locked place and/or secured with a safety device) were protective for suicide attempts among adolescents (Grossman et al., 2005}. A recent systematic review of clinic and community?based education and counseling around safe storage of firearms found that the provision of safety devices significantly increases safe firearm storage practices compared to counseling alone or providing economic incentives to acquire safety devices {Rowhani-Rahbar, Simonetti, Rivara, 2016). The Emergency Department Counseling on Access to Lethal Means (ED CALM) program trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post study, Runyan et al (2015) found that among the parents contacted at follow-up, 76% reported all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at the time of the child?s initial emergency department visit, all reported guns were currently locked, compared to 57% reporting this at the time of the initial visit. Organizational policies and culture Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components of Togetherfar Life were designed to foster an organizational culture that promoted mutual support and solidarity among members of the Force, help for problems related to suicide, training of supervisors, managers and all units to improve competencies in identifying suicidal risk and in using existing resources, and an education campaign to improve awareness and help?seeking. After implementation of the program, police suicides were tracked over 12 years and compared to rates in the control city of Quebec. Pre- post assessments of learning, interviews, and focus groups were also included. The suicide rate in the intervention group decreased significantly by 78.9% [p .008} to 6.42 per 100,000 per year compared to 29.0 per 100,000 in the control citya?a significant difference in rates {p (Mishara 8 Martin, 20121 The United States Air Force Suicide Prevention Program inclusive of 11 policy and education initiatives, was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training}, and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service-wide problems impacting the whole community. Using a time-series design, the AFSPP program 18 was associated with a 33% relative risk reduction in suicide [Knox et al., 2003). The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch] found significantly lower rates of suicide after the program was launched than before [Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years {Knox et al., 2010). Community-based policies to reduce excessive alcohol use Several studies on alcohol outlet density in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Giesbrecht et al., 2014; Escobedo 8i. Ortiz, 2002). For example, Giesbrecht et al. found that both on and off?premises alcohol outlets restaurants where alcohol is served and stores where alcohol is available for purchase to go) were positively associated with alcohol?related suicides in 14 U.S. states, particularly among men 1.08, and American Indiaanlaska Natives (ADE: 1.36; CI: Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 yea rs in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald 3t Remer, 2009). These findings suggest that measures to reduce alcohol outlet density can potentially reduce alcohol~involved suicides. 19 Promote Connected ness to Protect Against Suicide Rationale The ouantitv and qualitv of our social connection with others has been linked with suicide dating as far back to Durkheim, who first posited that the weakening social bonds is among the chief causes for suicidalitv?. Connectedness is the degree to which an individual or group of individuals are sociallv close, interrelated, or share resources with othersz. Connectedness can be formed within and between multiple levels of the social ecologv3; for instance between individuals leg. peers, neighbors, co? workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Connectedness is also viewed as a broad term that encompasses several other concepts including social support, social participation, social isolation, social cohesion, social capital, social integration, all of which have been linked to suicidal behaviors either conceptuallv or scientificallv within the literature. lconnectedness can serve either to protect or mitigate suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversitv. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primarv prevention activities to the community as a wholez. Insert info on connectedness and ind?udesome or all at the following re social capital here].50cial capital refers to a sense of trust in one?s communitv and neighborhood, social integration, and also the availabilityr and participation in social organizations (Muenning et al., 2013; Bever et al., 2014]. Manv ecological cross?sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generallv, and suicide. While the evidence is still being built the pattern is towards an inverse association between social capital measured bv social trust, communitv/neighborhood engagement, and improved mental healtl] Approaches The following two approaches are designed to promote connectedness among individuals and within communities to protect against suicide. Peer norm approaches seek to normalize help-seeking, encourage reaching out and talking to trusted adults, and promote supporting peers through building connectedness. These approaches are tvpicallv delivered in school settings but can also be implemented in community settingsComment IAI: This is what was in the table that Linda said could be added to the rationale. Community engagement activities. Community engagement is an aspect of social capital and involves residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes Reduction in maladaptive coping attitudes and behaviors Increase in healthy coping attitudes and behaviors Increase in referrals for youth in distressed Increase help-seeking behaviors Positive perception of adult support Evidence Brief statement to introduce the evidence. Current evidence provide some support for these types of approaches for reducing risk factors associated with suicidal behaviors. I Peer norm approaches. Evaluations show that programs such as Sources of Strength (505) can improve school norms and beliefs about suicide that are created and disseminated by student peers. in a randomized controlled trial of the program conducted with 18 urban and rural high- schoois Wyman et al., (2010fl found that $05 improved peer leaders' knowledge of adult supports for suicidal peers and were more likely than untrained leaders to refer distressed peers to adult supports Trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders I Community-engagement activities. Greening vacant urban spaces is an initiative that communities have adopted, particularly after the Great Recession (200?~2009) when many cities across the U.S. experienced urban abandonment. These initiatives engaged community members in the cleaning, greening, or beautifying vacant areas. One vacant lot greening initiative, which greened 4,436 lots in 4 areas of Philadelphia, PA, greening significantly reduced community residents' self-reported stress levels ip<.001} and engaged in more physical exercise ip<.001) than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism? 21 Teach Coping and Problem-Solving Skills Rationale Building life skills prepare individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem- solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors?. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theoriesf, surmising that individuals who engage in suicidal behavior is attributed to either direct learning, modeling, and environmental and individual hopelessness} characteristics. The literature linking life skills and suicide is The inability to employ adequate ceping strategies to cope with immediate stressors or identify and find solutions for problems have been characterized among suicide attemptersa. Treatments that include bolstering problem skills9 and include problem?solving techniquesmv11 appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents important life skills to offset the underlying vulnerabilities that contribute to engaging in high risk behaviors early in life. Approaches Current evidence provides support for the following two approaches: - Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high-risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work. I Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse. 22 Potential Outcomes I Reduction in suicide attempts and suicide ideation I Enhanced knowledge of risk and protective factors associated with suicide I Reduction in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) I Improve and normalize help-seeking behavior 0 Enhance social competence and emotional regulation skills 0 Enhance problem-solving and conflict management skills Evidence There are a several programs with evidence that supports teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Social emotional learning programs. These programs, commonly delivered in schools, overall have demonstrated success in preventing suicide risk. However, the evidence is mixed; while most models demonstrated strong effects, others were mixed. For example, Signs of Suicide is a high school-based prevention program for students designed to increase knowledge about suicide improve and normalize help-seeking behaviors. After 3 months of participation, 505 has documented a 54% reduction in suicide attempts among its participants compared to controls. 505 participants with a lifetime history of suicide attempt were also significantly less likely to report planning a suicide in the 3 months following the program compared to lower-risk participants. Moreover, increased knowledge of how access help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help-seeking were noted among students who participated in 505?. Another example is The Good Behavior Game {636), which is a classroom-based program for elementary school children uses a behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. In the evaluation of this program, several outcome behaviors were measured including suicide, given its association with externalizing behavior problems. Results from the first cohort evaluation found that first graders assigned to 686 reported half the adjusted odds of suicide ideation and suicide attempts than controls. After 14 years postvention, 636 was also found to significantly reduce risk of substance abuse?. However, in a replication trial, 636 did not produce the same reductions on suicide ideation or attempts as it did in the first cohort?. This finding likely arose due to the lack of implementation fidelity and pointed to the need for (336 to be delivered with precision, consistency, and teacher support 23 Finally, the Youth Aware of Mental Health Program (YAM), a school-based program rigorously evaluated in 10 European countries, teaches youth about suicide risk and protective factors and problem-solving skills. At 12 months, students participating in YAM were significantly less likely to have a suicide attempt and severe suicidal ideation compared to controls; Risk of suicide ideation and attempts fell by 49 to 54% among YAM participants?. Parenting skill and family relationship programs Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. One program, The incredible Years is a universally delivered program for parents of children up to 12 years of age designed to improve responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence and reduce problem behaviors in children. Several studies have demonstrated the effect of The incredible Years program on reducing internalizing such as anxiety and depression, and child conduct The program is also associated with improved problem-solving and conflict management, skills that were maintained at 1-yearfollow-up17'13'19. The program also demonstrated greater benefits as the dosage of the intervention increased?. Strengthening Families 10-14 years is a program that involves sessions between parents, youth, and family with the goals of: 1) improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; 2) promoting youths? interpersonal and problem-solving skills; and 3) creating family activities to build cohesion and positive parent-child interactions. The program has successfully shown to decrease externalizing behaviors, alcohol use, and drug use among youth participants and reductions in depression, alcohol use, and drug use among participating familiesu. 24 Identify and Support People At-Risk Rationale In order to be successful in decreasing suicidal behavior, attention must be paid to those who are at-risk or vulnerable. These persons experience risk and occurrence of suicidal behavior that is higher than average. This group requires particular focus on proactive case finding and retention and access to services. These vulnerable or disadvantaged populations include (but are not limited to): those living in lower socio-economic status; members of certain ethnic minority groups; those with a mental health problem; those who are institutionalized; those who have been victims of violence; and those who are homeless. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and maintaining care are still key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need ofthem. Nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches This document outlines three approaches that focus on identifying and supporting those who are at?risk- - Gatekeeper training is typically implemented in schools and within health care settings and is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care to assure that people who may be at high-risk of suicide don?t 'slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow-up. Programs such as these have demonstrated beneficial effects on depression, suicide ideation, and suicide mortality. In Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers andfor professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means restriction, 25 crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes - Reduction in suicide attempts 0 Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior I Increased at?risk individuals in treatment 0 Increased community members trained to identify at-risk individuals 0 Increased referrals for health care Evidence There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained. {Substance Abuse and Mental Health Services Administration, Center for Mental Health Services. Report to Congress: Garret Lee Smith Suicide Prevention Program. US Department of Health and Human Services: Rockville, MD, 2014). There is limited evidence for effectiveness screening programs but standard principles for public health screening make them promising. (Pena JB, Caine ED. Screening as an Approach for Adolescent Suicide Prevention. Suicide and Life-Threatening Behavior. 2006; The number of studies evaluating crisis intervention services is limited but a few studies do indicate that those who use the hotline services have decreased suicidal thoughts and behavior. Gatekeeper training. Mental Health First Aid (MHFA), designed for the lay public, consists of three weekly sessions of three hours each. The content covers helping people in mental health crises and/or in the early stages of mental health problems. The crisis situations covered included suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior. The mental health problems discussed included depressive, anxiety and disorders. The co-morbidity with substance use disorders is also covered. Participants learn the of these disorders, possible risk factors, where and how to get help and evidence-based effective help. In a randomized controlled trial of 300 participants, the intervention group reported greater confidence in providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. An additional finding was an improvement in the mental health of the participants themselves. All results were statistically significant at p<.05. {Kitchener El Jorm, 26 2004). Additional research rigorously evaluating MHFA for its impact on the first aid recipients themselves and suicidal behavior is needed (Kitchener 81 Jorm, 2006). Gorret Lee Smith (GL5) Suicide Prevention Program, as of June 2014, 154 GL5 grants had been awarded to 49 states and 48 tribes. Gatekeeper training has been a core part of all GL5 programs, and grantees have consistently reported spending the largest proportion of their budget on this strategy. A multi-site evaluation assessed the connection between community gatekeeper training (activities such as the number of people trained and the intended mid- and long-term outcomes) and a reduction of suicide attempts and deaths. To address this question, the analysis compared the change in the suicide mortality rates and nonfatal suicidal behavior amongthe population aged youth and young adults in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Effect on fatal behavior When compared with similar counties that did not implement GLS training, counties implementing GLS trainings presented significantly lower youth suicide rates the year following the training implementation (-1.07, This finding represents a decrease of 1 suicide death per 100,000 10 to 24 year olds or the avoidance of approximately 237 deaths in this age group between 200? and 2010. Effect on nonfatal behavior - Counties implementing GL5 program activities had significantly lower suicide attempt rates among youths 16 to 23 years of age in the year following implementation ofthe GL5 program than did similar counties that did not implement GL5 program activities (4.9 fewer attempts per 1000 youths [95% Cl, 1.8-8.0 fewer attempts per 1000 youths]; .003). More than 79 000 suicide attempts may have been averted during the period studied following implementation of the GL5 program. Walrath C, Godoy-Garraza L, Reid H, Goldston DB, McKeon R. impact of the Garrett Lee Smith Youth Suicide Prevention Program on Suicide Mortality. American Journal of Public Health: May 2015, Vol. 105, No.5, pp. 986-993. doi: Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 2015; 72:1143-1149. Brief statement introducing the evidence. 1- Screening combined with care management and overall continuity of care. Henry Ford Perfect Depression Core (Pre-cursor to Zero Suicide) The overall goal of the program was to eliminate suicide. More broadly, the aim of the program was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety. The redesign focused on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each ofthese aims. The program began with screening and assessment of each patient for suicide 27 risk with coordinated continuous follow-up care system wide. On the basis of the combined total of 31 suicides for the 11-year observation period, the rate of suicide among patients was 97 per 100,000 lN=13l for the two baseline years {the average rate for 1999 and 2000). This rate is similar to that reported for a clinical population. For the start-up year (2001), the rate of suicide was 41 per 100,000 For the follow?up interval {the average for 2002?2009], the rate was 19 per 100,000 Poisson regression analysis showed a statistically significant decrease of 82% in the suicide death rate between the baseline (1999?2000) and intervention (2002?2009) years (rate 95% confidence 535.001}. Coffey et al., 2013; Coffey, 2006. Additionally, between 1999 and 2010, researchers found that suicide rates declined among HMO members who received mental health specialty services, in association with a target prevention effort, and increased among HMO members who accessed general medical services but not specialty MH services (Coffey et. al., 2015]. Crisis intervention. Notionoi Suicide Prevention Lifeline (NSPL). This is a nationwide hotline that operates 24/? and is accessible by phone or a web-based chat function. Trained counselors are on-hand to listen, offer fee and confidential emotional support, and provide referrals for mental health services in the local area. Suicide callers lN=1,085) were assessed during their calls to the hotline and 380 participated in the follow-up assessment. Results indicated that seriously suicidal individuals called the hotline. Significant decreases in suicidality were found during the course of the telephone session, with continuing decreases in hopelessness and pain in the following weeks (Gould et al., 2007). Applied Suicide intervention Training This a training program for hotline counselors, emergency workers, clergy, caregivers and others in the community. The ASIST model has three phases of caregiving: connecting, understanding and assisting. The training helps participants identify people who are having thoughts of suicide and to recognize their invitation for help (connecting); to listen to the caller?s reasons for dying and living (understanding); and how to conduct a safety assessment, develop a safety plan for the person at risk, and connect the person at risk to community resources (assisting). The ASIST training program has been field tested in a variety of settings. in a national randomized controlled trial, Gould et al (2013) assessed the impact of the ASIST training across the NSPL network of hotlines over the period 2008-2009. Data were derived from 1,507 monitored calls from 1,410 suicidal individuals to 17 Lifeline centers. Callers who spoke with ASIST?trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of calls compared to those who spoke to counselors receiving usual care training. ASIST-trained counselors were also better able to keep callers on the phone longer and 28 establish a connection. ASIST training, however, did not yield more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 29 Intervene to Lessen Harms and Prevent Future Risk Rationale Individuals who have experienced mental health challenges, suicidal ideation, and have had non-fatal suicide attempts or have engaged in non-suicidal self-injury are at increased risk of subsequent suicide- related morbidity and mortality. Risk of suicidality can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide. Exposure to sensationalized or uninformed reporting regarding suicide-related deaths may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Approaches A broad array of approaches to lesson harms and reduce future risk of suicidality among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion. I Treatment for people at-risk of suicide typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. I Treatment to prevent re?attempts. These follow-up contact approaches use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors to prevent reattempts. These approaches typically focus on coping and other emotional regulation skills and may include case management home visits to increase adherence to and continuity of care, one?on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%-25% reattem pt within a year, and of attempt survivors die by suicide within 1 to 5 years oftheir initial attempt (lnagaki, et al., 2015}. I Postvention approaches are implemented after a suicide has taken place and may include debriefing of survivors [those who have lost a friend, peer, family member, co-worker to suicide}, counseling, and/'or bereavement support groups. The programs have not typically tested their impact on suicide or suicidal behavior but may reduce risk of guilt, feelings of depression, and complicated grief {Szumilas et al., 2011). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (no outreach) {Cerel 8: Campbell, 2003i 30 Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Potential Outcomes I Reduction in mental health-related sequelae Increase I Improved coping skills 0 Improved messaging following suicide 0 Reduction in re-attempts Evidence The evidence addressing strategies to lesson harm and prevent future risk ofsuicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation of suicide-related mortality is a statistically rare event, evaluation of mortality outcomes requires large sample sizes and extended follow-up. Therefore, much of the evidence in this area primarily focuses on rislt and protective factors. Treatment for people at-risk of suicide. has been demonstrated to help diverse individuals in different settings to reduce risks related to mental health problems such as depression and anxiety, and to reduce suicidal ideation and attempts. Evaluation evidence for selected approaches to prevent future suicide risk follow below. Improving Mood Promoting Access to Collaborative Treatment (IMPACT) aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. IMPACT facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase) by a depression care manager. IMPACT has been shown to significantly improve quality of life, and to reduce functional impairment, depression (Hunkeler et al., 2006; Uniitzer et al., 2002) and suicidal ideation over 24smonths of follow-up (Uniitzer et al., 2006} relative to patients who received care as usual. 31 Collaborative Assessment and Management of Suicidality (CAMS) is a therapeutic framework for suicide-specific assessment and treatment of patient?s suicide risk. it is a flexible approach that can be used across treatment settings and clinician theoretical orientations. The clinician and patient work together in an interactive assessment process. The patient is highly engaged in the development of their own treatment plan. Every session of CAMS is collaborative and involves the patient?s input about what is and is not working. Ultimately, this process is designed to enhance the therapeutic alliance and increase treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies (lobes, 2012), in a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community- based sample of suicidal outpatients. (Comtois et al., 2011). Dialectical Behavioral Therapy is a multicomponent therapy for individuals at high risk for suicide who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self-injurious behavior, those receiving DBT were half as likely to make a suicide attempt at twonyear follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self?injurious acts combined (Linehan et al., 2006). Attachment-Based Family Therapy (ABFT) is a program for adolescents ages 12-18 that is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety. A randomized controlled trial of ABFT (Diamond et al., 2010} found that suicidal adolescents assigned to ABFT experienced significantly greater improvement regarding suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care (-4.37 vs. -2.34; .001; d=0.97l. A higher percentage of ABFT participants reported no suicidal ideation in the past week at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%; .01; OR: 4.25) and at 24 weeks (82.1% vs. 46.2%; .006; Treatment to prevent re?attempts. Active contact and follow-up approaches intended to prevent reattempts among patients that have been hospitalized and subsequently discharged for suicide attempts have been found in a meta-analysis conducted by lnagaki et al., {2015) to reduce reattempts by approximately 17% for up to 12 months post-discharge, however, the long-term effects of these approaches on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect 32 of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of postcrisis suicide prevention long-term follow-up contact approach, Motto et al., [2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of followup than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam et al., 2011; Wang et al., 2016]. I Postvention approaches such as StondBy Response Service provide suicide bereavement support services to clients via face~to~face outreach and telephone support delivered by a professional crisis response team. A site coordinator develops a customized case management plan, referring clients to other existing community services matched to their needs. In a study by Visser et al. (2014], StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and Edi/E. respectively, 0.005). 0 Safe messaging following a suicide. Media guidelines for reporting on suicides can help assure that stories on suicide are communicated in a safe and effective way. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrUpted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a reduction of 81 suicides annually (95% confidence interval: -149 to ?2.32, df 54, 0.024] in the Viennese subway system [Niederkrotenthalen T., Sonneck, (3., 2007) Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population?level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business/labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non?governmental organizations. Collectively, these sectors can 33 make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and in?uential role 34 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation; planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certificates, law Enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair et al., 2016). The National Electronic Injury Surveillance System-All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.}, age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as school-based state, territorial, tribal, and large urban school district survey conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual nationwide survey of individuals aged 12 years and older that provides national and state-level estimates of drug use and mental health-related issues, including suicide ideation and suicide attempts It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. 35 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation. However, there is potential to understand the synergistic effects within a comprehensive prevention approach. Additional research is needed to understand the extent to which combinations of strategies and approaches result in greater reductions in suicide than individual programs, practices, or policies. Conclusion 36 References 37 Appendix A: Summary of Strategies and Approaches to Prevent uicid St rategv Approach} Program, Practice or Policy Strengthen ?nancial security Best Available Evidence Suicide Suicide Other Attempts or Risk! Protective Ideation Factors for Suicide Lead Sectors? 1 Comment help you complete the lead sector column. For the other columns, you just need to insert a check? mark based on the evidence vou describe In the narrative for a particular program or policy. For example, if the evidence shows impact on suicide, then put a check-mark in that column. If the studvr also found effects on risk or protective factors, then put a check-mark in that Strengthen Unemployment benefit programs economic supports Housing stabilization policies The National Neighborhood Program Strengthen Coverage of mental health conditions in health insurance policies BCCESS to mental health care Mentoi Heol'th Parity Lows Establish protective environments Means restriction intervening at hot spots Safe storage practices it (attempts) it [Safe storage of ?rearms and med ication) Organizational policies and culture Tag ether for Life US Air Fa rce Suicide Prevention Pro gram [familv violence) Lcolumn as well. Best Available Evidence Communitywbasecl policies to reduce excessive alcohol use Aicohoi outiet density Peer norm approaches Promote connectedness Sources of Strength Ni to protect against suicide Communitvrengagement activities Greening vacant urban spaces ?i Social emotional learning Youth Aware of Mentai Heaith Program xi Signs of Suicide *1 Teach co in 3 Good Behavior Game Ni and problem- Parenting skill and family relationship approaches The incredibie Years xi Strengthening 10-14 Gatekeeper training Mentai Health First Aid Best Available Evidence Screening combined with care management Identify and support people Henry Ford Perfect Depression Care {Pre- at-risk cursor to Zero Suicide) Crisis Intervention Nationai Suicide Prevention Lifeiine Appiied Suicide intervention Training Treatment for people at risk of suicide improving Mood - Promoting Access to Caiiabarative Treatment Coiiaharative Assessment and Management Intervene to ofSuicidaiity (CAMS) lessen harms and prevent future risk Diaiecticai Behaviorai Therapy Attachment-Based Famiiy Therapy Treatment to prevent re-attempts ED Brief intervention with Foiiow-up Visits Active foiiaw?up con tact approaches CBTfor Suicide Prevention Best Available Evidence Postvention StandBy Response Service Safe messaging following a suicide Media Guidelines - 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. 1 Butchart A, Phinney A, Check P, Villaveces A. Preventing violence: a guide to implementing the recommendations of the World Report on Violence and Health. Geneva, Switzerland: World Health Drganizaiton; 2004. 1 Klevens J, Simon TR, Chen J. Are the perpetrators of violence one and the same? Exploring the co?occurrence of perpetration of physical aggression in the United States. interpers Violence. 3 Desai, R. A., Dausey. D. J.. Rosenheck, R. A. (2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. American Journal of Coulton CJ, Crampton DS, lrwin M, Spilsbury JC, Korbin JE. How neighborhoods influence child maltreatment: a review ofthe literature and alternative pathways. Child Abuse Negl. 5 Freisthler B, Merritt DH, LaScala EA. Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat. 5 Capaldi DM, Knoble NB, Shortt JW, Kim HK. A systematic review ofrisk factors for Intimate partner violence. Partner Abuse. Pinchevsky GM, Wright EM. The impact of neighborhoods on intimate partner violence and victimization. Trauma Violence Abuse. 3 Sampson RJ, Morenoff JD, Gannon-Rowley T. Assessing "neighborhood effects": Social processes and new directions in research. Annual review of sociology. 9 Luo F, Florence CS, Quispe-Agnoli M, Ouyang L, Crosby AE. impact of business cycles on US suicide rates, 1923-2007. American journal of public health. 1? Reeves A, Stuckler D, McKee M, Gunnell D, Chang SS, Basu 5. Increase in state suicide rates in the USA during economic recession. The Lancet. 1814. 11 Runyan D, Wattam, C, lkeda, FL, Hassan, F, Ramiro, L. Child abuse and neglect by parents and caregivers. In: Krug E, Dahlberg LL, Mercy, JA, Zwi AB, Lozano Ft, ed. World report on violence and health. Geneva, Switzerland: World Health Organization; 200259-86. 11 Heise L, Garcia-Moreno C. Violence by intimate partners. In: In: Krug E, Dahlberg LL, Mercy, JA, Zwi AB, Lozano R, ed. World report on violence and health. Geneva, Switzerland: World Health Organization; 2002:87-121 13 Pinchevsky GM, Wright EM. The impact of neighborhoods on intimate partner violence and victimization. Trauma Violence Abuse. 1? Centers for Disease Control and Prevention, Sexual Violence: Risk and Protective Factors. Available at: Accessed April 4, 2014. 15 Wilson WJ. When work disappears: The world of the new urban poor. New York, NY: Random House 2011. 15 Briere, J., Madni, L. A., Godbout, N. (2015). Recent Suicidality in the General Population Multivariate Association With Childhood Maltreatment and Adult Victimization. Journal ofinterpersonal violence, 0886260515584339. 17 Logan IE, Lceb RT, Barker LE. Gender-speci?c mental and behavioral outcomes among physically abused high-risk seventh-grade youths. Public Health Reports. 13 Duke NN, Pettingell SL, McMorris Bl, Borowskv IW. Adolescent violence perpetration: associations with multiple types of adverse childhood experiences. Pediatrics. 19 Duke NN, Pettingell SL, Bl, Borowskv IW. Adolescent violence perpetration: associations with multiple types of adverse childhood experiences. Pediatrics. 2? Violence and the Family: Report oftlte American Association Presidential Task Force on Violence and the Family. Washington. DC: American Association; I996. 2? Violence and tire amily.? Report oftlze American Association Presidential Task area on Violence and the Family. Washington, DC: American Association: 1996. 22 Violence and the Family: Report oftlie American Association Presidential Taslc Force on Violence antir the Family. Washington, DC: American Association; 1996. 23 Jewkes R. Rape perpetration: a review. Pretoria, Sexual Violence Research initiative. 20I 2. 1" Coulton. C. J., Crampton, D. 8., Irwin, M., Spilsbury, (3., 8; Korbin. J. E. (2007). How neighborhoods influence child maltreatment: A review of the literature and alternative pathways. Child abuse ll. neglect, 31(11), 11 174 142. 25 Pinchevsky GM, Wright EM. The impact of neighborhoods on intimate partner violence and victimization. Trauma Violence Abuse. Ill 32. 2?5 Basile KC, Hamburger ME, Swahn MH, Choi C. Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. The Western Journal of Emergency Medicine. 2013;14l411329-340. 27 Widorne R, Sieving RE, Harpin SA, Hearst MO. Measuring neighborhood connection and the association with violence in young adolescents. Journal of Adolescent Health. 2?3 Kleiman, E. M, Riskind. J. H., Sohaefer, K. E., St Weingarden. H. (2012). The moderating role of social support on the relationship between impulsivity and suicide 33t?lz273?2l9 25? Centers for Disease Control and Prevention, Elder Maltreatment: Risk and Protective Factors. Available at: Accessed April 4, 2014. 3? Borowskv IW, Hogan M, Ireland M. Adolescent sexual aggression: risk and protective factors. Pediatrics. 3? Capaldi DM, Knohle NB, Shortt .IW, Kim I-IK. A systematic review of risk factors for Intimate partner violence. Partner Abuse. 32 DcGue S, Massetti GM, Holt MK, et al. Identifying links between sexual violence and youth violence perpetration. of Violence. 140-156. 33 Basile KC, Espelage DL, Rivers 1, McMahon PM, Simon TR. The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and Violent Behavior. 2009; 34 Losel F. Fartington DP. Direot protective and hollering protective factors in the development of youth violence. American Journal of Preventive Medicine. 20] 35 Hong S, Espelage DL. A review of research on bullying and peer victimization in school: An ecological system analysis. Aggression and Violent Behavior. 3'5 Carter, M, McGee, R., Taylor, 8., Williams, S. (2007). Health outcomes in adolescence: Associations with family, friends and school engagement. Journal ofaa?olescence, 30(1), 51-62. 3'7 Centers for Disease Control and Prevention. Child Maltreatment: Risk and Protective Factors. Available at: Accessed April 4, 2014. 3'3 Capaldi DM, Knoble NB, Shortt JW, Kim HK. A systematic review of risk factors for Intimate partner violence. Partner Abuse. 39 Resnick, M. D., Ireland, M, 8. Borowsky, I. (2004). Youth violence perpetration: whet protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. Journal of adolescent health. 35(5), 424-e1, ?0 Elgar FJ, Craig W, Boyce W, Morgan A, VeIIa-Zarb R. Income inequality and school bullying: multilevel study of adolescents in 37 countries. Journal of Adolescent Health. Maimon. D., Browning, C. R, Ex Brooks-Gunn, J. (2010). Collective ef?cacy, family attachment, and urban adolescent suicide attempts. Journal of health and social behavior, 51 42 Capaldi DM, Knoble NB, Shortt W, Kim HE. A systematic review of risk factors for Intimate partner violence. Partner Abuse. ?3 Losel F, Farrington DP. Direct protective and buffering protective factors in the development of youth violence. American Journal of Preventive Medicine. ?4 Maimon, D., Browning, C. R.. 8: Brooks-Gum. J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Journal of health and social behavior, 51(3), BOT-324. ?5 Capaldi DM, Knobel NB, Shortt JW, Kim HK. A systematic review of risk factors for intimate partner violence. Partner Abuse. 45 Ltisel F, Farrington DP. Direct protective and buffering protective factors in the development of youth violence. American Journal of Preventive Medicine. References for Shane?s section of risk and protective factors: 1- Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56 2' World Health Organization. World Health Organization. Geneva, Switzerland: World Health Organization, 2014. 3- U.S. Department of Health and Human Services Office of the Surgeon General and National Action Alliance for Suicide Prevention. National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: HHS, September 2012. Zalsman G, Hawton K, Wasserman D, et al. Suicide prevention strategies revisited: 10-year systematic review. Lancer Ref: Blair JM. Fowler KA, Jack SPD, Crosby AE. The National Violent Death Reporting System: overview and future directions. in) Prev 2016;22:8uppl 1 iB?i11. NEISS AIP Ref: CDC, Web-based Injury Statistics Query and Reporting System, Nonfatal Injury Reports, 2001-2014. Available at: Shane?s section: 1. 2. 10. ll. Durkheim E. Suicide: A study in sociology. Glencoe Press: New York; 1951. Centers for Disease Control and Prevention (CDC) (2011). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Oahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56 Wyman, P. A., Brown, C. H., LoMurray, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Pubiic Heoith, 100(9), 1653-1661. Branas, C. C., Cheney, R. A., MacDonald, l. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011}. A difference-in-differences analysis of health, safety, and greening vacant urban space. American Journal of Epidemiology, 174(11), 1296-1306. World Health Organization. World Health Organization. Geneva, Switzerland: World Health Organization, 2014. Bandura A (1986) Social Foundations of Thought and Action: ASocial Cognitive Theory. Englewood Cliffs: NewJersey: Prentice-Hall Publishers. Pollock, LR. and Williams, MG. (2004}. Problem-solving in suicide attempters. Medicine, pp 163-167. doi:10.101750033291703008092. Goldsmith, S. K., Pellmar, T. C., Kleinma n, A. M., Bunney, W. E. (2002). Reducing suicide: A notionoi imperative: National Academies Press. Townsend, E., Hawton, K., Altman, 0., Arensman, E., Gunnell, D., Hazell, P., . . . Van Heeringen, K. (2001). The efficacy of problem- solving treatments after deliberate self-ha rm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. medicine, 31(6), 979-988. M. Ghahramanlou-Holloway, S. S. Bhar, G. K. Brown, C. Olsen and A. T. Beck (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention ofsuicide. Medicine, 42, pp 1185-1193. doi:10.101750033291711002169. 12. 13. 14. 15. 16. 18. 19. 20. 21. Schilling, E. A., Aseltine, R. H., Jr., James, A. (2016). The 505 Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-156. Kellam, S. (3., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., Toyinbo, P., et al. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug and Alcohol Dependence, 95(Suppl. 1), 55-528. H.C. Wilcox, S.G. Kellam, C.H. Brown, J. Poduska, N.S. Ialongo, W. Wang, J. Anthony. The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 (2008), pp. 560? $73 Wasserman, 0., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, 6., . . . Carli, V. (2015). School-based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. doi: 10.1016150140- Webster-Stratton, C., Reid, M. J., 8; Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: Evaluation of the Incredible Years teacher and child training programs in high-risk schools. Journal of Child and 49(5), 471-488. Reid, M. J., Webster-Stratton, C., Hammond, M. (2003). Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34, 471-491. Webster-Stratton, C., 8; Hammond, M. (1997). Treating children with early-onset conduct problems: A comparison of child and parenting training interventions. Journal of Consulting and Clinical 65(1), 93-109. Webster-Stratton, (2., Reid, M. J., El Hammond, M. (2001). Preventing conduct problems, promoting social competence: A parent and teacher training partnership in Head Start. Journal of Clinical Child 30(3), 233-302. Herman, K. C., Borden, L., Reinke, W. M, 8: Webster- Stratton, C. (2011). The impact of the Incredible Years Parent, Child, and Teacher Training Programs on children's co-occurring internalizing School Quarterly, 26, 189?201. doi:10.1037/ 80025223. Spoth, R., Guyll, M., Day, 5. X. (2002). Universal fa min-focused interventions in alcohol-use disorder prevention: Cost-effectiveness and cost-benefit analyses of two interventions. Journal of Studies on Alcohol, 63(2), 219-228. Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 Policies, Programs, and Practices to Support individuals, Families, Communities: A Technical Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technicai Package to Prevent Suicide. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Strengthen Access to Mental Health Care 16 Create Protective Environments 18 Promote Connectedness 22 Teach Coping and Problem-Solving Skills 24 Identity and Support People At-Risk 28 Intervene to Lessen Harms and Prevent Future Risk 32 Sector Involvement 37 Monitoring and Evaluation 39 Conclusion 40 References 42 Appendix A 50 Acknowledgments [to be inserted later] External Reviewers [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called seif?directeo? vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, 81 Melanson, 2011). Self- directed violence may be suicide! or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fatal self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. in 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U5 (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, 8; Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by raceiethnicity with the highest rates, across the lifespan, occurring among non-Hispanic American lndiaanlaska Native (rate: Among young people are disproportionately at increased risk of suicide with young males aged 25-29 experiencing the highest rates (rate: 5697100000}. Moreover, suicide rates among non-Hispanic AIIAN have increased by 48.7% since 1999 (Centers for Disease Control and Prevention, 2016d). Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, 8t Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press}. Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological mode ?- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002}. Risk and protective factors for suicide exist at each level. For example, risk factors include: individual level: History of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization?d perpetration, and genetic and biological determinants Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General at National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). It is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8; Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, 8L Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, Irwin, Spilsbury, 8i. Korbin, 2007; Freisthler, Merritt, 8: LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8: Kim, 2012), intimate partner violence (Pinchevsky 8: Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, 8-: Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 81 Ramiro, 2002), intimate partner violence (Heise 8: Garcia-Moreno, 2002; Pinchevsky Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, 8 emmcomeesdnee to err-?5 Borowsky, Hogan, 8: Ireland, 199?; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, 8t Weingarden, 2012; Pinchevsky Wright, 2012; Widome, Sieving, Harpin, 8t Hearst, 2003), school (Basile, Espelage, Rivers, McMahon, 8a. Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, 0 Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, SLAIIen?Meares, 2012; Losel Fa rrington, 2012), family [Ca paldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, 8; ?v'ellavZarb, 2009; Maimon, Browning, 81 Brooks-Gunn, 2010; Resnick, Ireland, at Borowsky, 2004), caring adults (Capaldi et al., 2012; Losel 8: Farrington, 2012; Maimon et al., 2010), and pro-social peers (Capaldi et al., 2012; Losel 8: Farrington, 2012) enhances resilience to suicide and other forms of violence. .- Comment IAI: From TS: lfound this wording somewhat awkward. Comment IA): From LLD: The Crosby 8: Sacks reference is missing from the reference list. i looked it up onllne and noticed that the 13 million estimate is based on knowing a suicide decedent in the past year. Knowing someone who died by suicide seems potentially broader to me than being directly impacted. Comment IA): From TS: Has anyone updated this?I It would be good to check. 1 Comment IAI: From LLD: they may also be impacted in other ways loss of The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, El. Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015]. age .lnanearly II I I study, Crosby and Sacks l2002) estimated that 2% of the U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et. al {2016) found that 48% of the weighted survey population knew at least one person who died by suicide in their lifetimes. Research indicates that the impact of knowing someone who died by suicide andlor having lived experience li.e. having attempted suicide oneself) is much more extensive than injury and death. Peopliwith lived experience may suffer long?term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt {Chapman Dixon?Gordon, 200?). Similarly, survivors of a loved one?s suicide may EaxperiencelongoIn-g?i pain and suffering including complicated grief (Mitchell, Kim, Prigerson, El. Mortimer-Stephens, 2004), stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide Ilulie Cerel, McIntosh, Neimeyer, Maple, 8f. Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2008). L_ess discussed but no less important, are the financial and occggational effects for those left behind (Florence, Simon, Haegerich, Luo, a Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs ?orence et al., 2015). By another stimate flihe total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence in 2013 were approximately $93.5 billion after adjusting for undersreporting of suicide {Shepard, Gurewich, Lwin, Reed, 8t Silverman, 2015)] The overwhelming burden of these costs results Lfamily income, etc.) Comment From TS: it seems like it would be appropriate to acknowledge the CDC estimate too. Consider using that as a minimum and then say that others have estimated that the costs could be considerably higher after adjusting for ?nder-reporting ofsuicide. from lost productivityr over the life course, with the average cost per suicide being over $1.3 million {Shepard et al., 2016}. Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies exists, and manv programs are readv to be implemented now. Just as suicide is not caused bv a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting {Silverman 8i Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual, relationship, familv, communitv, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor] {National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014]. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or program, practice, or policv selected had to meet at least one of these criteria: a] meta-analvses or svstematic reviews showing impact on suicide; b} evidence from at least one rigorous le.g., randomized controlled trial or quasi-experimental design) evaluation studv that found significant preventive effects on suicide; cl meta~analvses or svstematic reviews showing impact on risk or protective factors for suicide, or cl] evidence from at least one rigorous RCT or quasi-experimental design) evaluation studv that found significant impacts on risk or protective factors for suicide. Finallv, consideration was also given to the likelihood of achieving bene?cial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibilltv of implementation in a U.S. context if the program, policy, or practice has been evaluated in another countrv. Within this technical package, some approaches do not vet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide helpnseeking, stigma reduction, depression, connectedness). In terms of the strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of communitv engagement and familv programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect i theEdeveloangmenta-lnature of the evidence base and the use of the best available evidence at a givenl time. It is also important to note that there is often significant heterogeneitv among the programs, policies, or practices that fall within one approach or strategv area in terms of the nature and qualitv ofthe available risk or protective factors for suicide. To be considered for inclusion in the technical packageComment IAI: From TS: ?developmental nature"I doesn't seem correct. Do vou I mean the ?current status? evidence. Not all programs, policies, or practices that utilize the same approach gatekeeper training) are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports u. Strengthen household financial security I Housing stabilization policies Strengthen access to mental health care i Coverage of mental health conditions in health insurance policies Create protective environments - Reducing access to lethal means among persons at-risk of suicide - Organizational policies and culture I Community-based policies to reduce excessive alcohol use Promote connectedness I Peer norm approaches I Community engagement activities Teach coping and problem?solving skills - Socialvemotional learning - Parenting skill and family relationship approaches Identify and support people at risk Iv Gatekeeper training I Screening combined with care management 11 Crisis intervention Treatment for people at?risk of suicide Treatment to prevent re-attempts Postvention Safe messaging following a suicide 0 Intervene to lessen harms and prevent future risk Oil! The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support may also vary by population and community characteristics. ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator le.g., strategies addressing community and societal level risks], but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 ?5 Strengthen Economic Supports -n Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015). Economic and ?nancial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress, can directly increase I an individual?s risk for suicide or indirectly increase risk by exacerbating related physical and mental hiajth groblems; buffering these risks can therefore, potentially protect against suicide (Stack 8i. Wasserman, 200?}. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. - Strengthening household ?nancial security can potentially bufferthe risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples ofways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 1 Comment IAI: From TS: was thinking that it might be good to work in a point about how these risks can be directly or indirectly associated with suicide risk because these stressors can exacerbate relationship as well physical and mental health problems. '93 .1 1 {Comment From Consistent tense reduced vs. reduce Comment IM: From LLD: the other packages Use ?RedIJctions you may want to consider Using similar language 2 for your potential outcomes. consistent throughblit. =lcze: -q nun?uh.- Int-AI-uun-Inq-nnunnut Ill-Ill.- a] Potential LutcomeJ Headsets-Reductions in suicide rates I Lower?Reductions in foreclosure rates - Reductions intewer eviction rates Red-aeed?Reductions in emotional distress 13 Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. An examination of variations in U.S. unemployment benefit programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits (Cylus. Glymour, Er Avendano, 2014]. Another U.S. study examining the link between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed tojuat the loss of job, predicted suicide risk {Classen 3.: Dunn, 2012]. Together, these results suggest that not only should state unemployment benefit programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen hOUSehold financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance] have also shown an impact on suicide. A study by Flavin and Radcliff [2009) examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance {Temporary Assistance to Needy Families and total state spending on suicide rates between 19902000, controlling for a number of suicide risk factors leg, residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn't spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin 8t Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased theirit_s per capita spending on these types ofare needed to further understand the outcomes impacted by programs such as these. 0 Housing stabilization policies. The Notional Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle?income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are 14 Comment IAI: From TS: Consider including a point about this being a correlational study and more evaluation work is needed but it suggests the potential benefits of policies that reach those who are particularly Vulnerable at the times when thEy are in greatest need. Needed. lea as: sets indicate Ibis risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014}. Another study of data from 16 US. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began) to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access to Mental Health Care 'c Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Jf I arias [ArsenauIt?Lapierre, Kim, &Turecki, 2004; E. C. Harris 81 Barraclough, 1997; mental illnesses diffeek Tvrer, Reed, 8: Crawford, 2015], previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 8: Barraclough, 1993; World Health Organization, 2014]. Studies suggest that up i I 1 time of their deaths (Arsenault?Lapierre et al., 2004; Cavanagh, Carson, Sharpe, 3: Lawrie, 2003; Isometsa, 2001}. State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Sherri, Bagalman, 8: Cao, 200?]. Findings from the National Comorbiditv Survev indicate that relatively few people in the 0.5. with mental health disorders receive treatment for those conditions (Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services [Cunningham 2009}. identifying wavs to improve access to timely, affordable, and qualitv mental health ca re for people I means that different tvpes of mental health problems pose different risks for suicide, but wasn't sure? Mav want to make the point more explicit Huang, 8: Stroup, 2015; Owens, 2002), End the level of riskeisk conferred bv different types of ?Hep added somelanguage than sweepers 1 Comment From TS: Isn't this to 90% of people who die by suicide mav have had a mental illnessior substance abuse problems at the ,res - Added. 1 omment IAI: From JM: assume this including substance abuse disorder? lt would be good to state this explicitlv. In. Comment IAI: From T5: lfound this sentence confusing. Can vou reword to this. From LLD: could sav: "if a state has a weaker paritv law than the federal paritv law leg, includes coverage for some mental health conditions but not others], then the federal paritv law will replace the state law.? in need is a critical component to suicide prevention {World Health Organization, 2014}. Apart from the treatment benefits, it can also serve to normalize help?seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns li.e., mental health paritv}. Benefits and services covered include such things as the number of visits, co-pavs, deductibles, inpatient/outpatient services, prescription drugs. and hospitalizations. If a state has a stronger mental health paritv law than the federal parit?,r law, then insurance plans regulated bv the state must follow the state pa ritv law. If a state has a weaker paritv law than the federal paritv law ie.g.,i includes coverage for some mental health conditions but not others], then the federal paritv law will replace the state law. Equal coverage does not necessarily implv good coverage as health insurance plans vars,r in the extent to which bene?ts and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. 15 Jhanks, Linda. Accepted your suggestion._, Potential Outcomes teeseased?lncreases in utilization of mental health services DecreasedReductions in of mental illnesses Deereased?Reductions in rates of suicide attempts Beereased-Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may:F directlv impact suicide rates. I Coverage of mental health conditions in health insurance policies. The National Survev of Drug Use and Health is a nationallv representative survev of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and See {2006) found that 12 months after states enacted mentoi health parity iows, self-reported use of mental healthcare services significantlv increased. Moreover, subsequent research by Lang {2013] examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health paritv laws, specificallv, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per vear._ Lang et. al (2013} estimated the cost of saving a life through such mandates as mental health paritv bv comparing the loss in wagg attributable to the policy, via increased premiums, to the number of lives saved. Based on these calculations, the cost ofsaving one life was between WEE-3.1 prevented. However, this calculation did not take into account the saving associated with improved mental health among non?suicidal individuals, increases in productivltv, or qualitv of life associated with increased mental health care. As such, this figure is considered the upper bounds on costs incurred and should be interpreted with caution (Lang, 2013}. 1 Comment IAI: From TS: i don?t think this is accurate. I was curious about this study and checked out the paper. I think they are saving that this is the cost incurred per suicide prevented. Thev go on to explain that this is the upper bound and does not reflect the benefits to non? suicidal individuals. Please con?rm this. If vou are going to include this then it will Lhe important to explain it further. 1? Create Protective Environments Rationale Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddom 1980}. Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2016; et al., 2016), therefore, changes to the organizational culture in these occupations, by way of implementation of supportive policies or even physical modifications to the workplace environment, can change social norms, encourage helprseeking, and demonstrate that good health and mental health are valued and Comment IAI: From lid: The word promising has certain connotations when using it around evidence. li?ou say later that the evidence around this strategy is some of the strongest for suicide. These two statements seem somewhat inconsistent. i would use a different word here then promising if you truly believe there is strong evidence. From JH: suggest deleting the word ?promising? From LLD: since we don't want to infer that these are the only approaches for creating protective environments, I might be inclined to change the word ?promising" to ?potential? or "poSsihle" or something along those lines. that stigma and other risk factors for suicide are not L. Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis (Beautrais, Gibb, Fergusson, Horwood, 31. Larkin, 2009,- A.E. Crosby, Espitia~ Hardeman, Ortega, 8t Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i Azrael, 2015; C. W. Runyan et al., 2016Approaches The current evidence suggests three Wootential approaches for creating environments that; protect against suicide. Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that 1) the interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001} and 2) that pe0ple tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, 200?; Yip et al., 2012]. Therefore, increasing the time interval between the thought and the suicide attempt, for example, by Thanliyou forthe suggestions. making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures leg, bridges and cliffs), railway tracks, and 18 isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable and/or impulsive from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition-?and keeping medicines in a locked cabinet or other secure location awayfrom people who may be at risk or who have made prior attempts (Rowhani-Rahbar, Simonetti, Rivara, 2015; C. W. Runyan et al., 2015L Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior asking for help}, skill building, changing social norms, referral and access to helping services leg. mental health, substance abuse treatment, financial counseling), and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (National Action Alliance for Suicide Prevention Workplace Task Force, 2015]. Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo 8: Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, Wilcox, 2004). Potential Outcomes Increaseg in safe storage of means Reductions in suicide attempts Reductions in suicide deaths Increaseg in help-seeking Reductiong in alcohol-related suicide deaths Evidence The evidence for the effectiveness of preventing suicide by reducing access to lethal means feepersee at-eisk?ef?suieideand etheewaysrte otherwise establishi_ng protective environments for individuals at 19 5;ng suicide js_str9ng, particularly compared to existing evidence for other prevention strategies-is (Zalsman et al., 2016}. Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the 0.5. and abroad, found associated reduced rates of suicide (Cox et al., 2013,- Pirkis et al., 2015). For example, after erecting a barrier on the Jacques?Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year {Perron, Burrows, Fournier, Perron, 8: lDuellet, 2013]. Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites [Perron et al., 2013}. Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold [Beautrais, 2001; Beautrais et al., 2009]. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm?related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. [2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own {Rowhani-Rahbar et al., 2016]. Another program, The Emergency Department Counseiing on Access to Lethai Means CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 13 receiving care for suicidal behavior. In a pre?post quality improvement project, Runyan et al (2016} found that at post?test (of the 55% of parents followed up, n=114] reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post?test (C. W. Runyan et al., 2016]. Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among of?cers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and 20 Comment IAI: From .Il'v?l: Consistent with promising used earlier? Is it the strongest? I think it is consistent with promising, but We changed that to potential. The interpretation from the paper states no single strategy clearly stands above the others", so I would be hesitant to call it out as the strongest evidence. but it de?nitely may be for certain populations. Modified to simply $95.55!? 915% it is. Eli?En??a all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 8: Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 100,000) (Mishara 8: Martin, 2012}. Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement}. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 3: Caine, 2003). Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively}, homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997' launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010}. - Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 531 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8: Remer, 2009}. Promote Connectedness 21 . Comment IAI: From JM: Seems like a nuance you can delete from this paragraph. You don?t really pick up on this later and it probably requires more Rationale Sociologist, Emile Durkheim theorized in 139? that weak social bonds, i.e. lack of connectedness, are . among the chief causes for suicidality (Durkheim, Connectedness is the degree to which an I individual or group of individuals are socially close, interrelated, or share resources with others (Centers Lexplanation for Disease Control and Prevention, 2009). ocial connections can be formed within and between 5? rCummentlAl: FromJH,Awkwarg multiple levels ofthe social ecology (Dahlberg Bi Krug, 2002}, for instance between individuals leg. peers, I, CONS-idem while the evidence limited, existing studies neighbors, co?workers], families, schools, neighborhoods, workplace, faith comm unities, cultural groups, ,r and society as a whole] Related to connectedness, social capital refers to a sense of trust in one?s ,l cumulul If?: . 1. that prosociai behavior is currently community and neighborhood, social integration, and also the availability and participation In somal abmmai_ Maybe promuteisabemr From JH: suggest deleting ?seek to organizations (Beyer, Layde, Hamberger, 8t Laud, 2015; Muennig, Cohen, Palmer, 8; Zhu, 2013). Many ecological cross?sectional and longitudinal studies have examined the impact of aspects of social capital i normalize? and say "Peer norm i approaches encourage prosocial JL. word? 1- i I on depression depressive disorder, mental health more generally, and suicide. While the evidence is still?beingbul-Itlimited, existing studies sygge_st]_the pattern is towards a?s?t?emveese association between social capital measured by social and improved mental health. Connectedness and social capital together can serve to protect against behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of Comment I prefer to keep it as normalize and took out prosocial behavior. We want to normalize help- seeking versus making it seem like a Lpersonal weakness. suicidal adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole [Centers for Disease Control and Prevention, 2009]. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. Peer norm approaches seek?o factors for suicide such as help?seeking, reaching out and talking to trusted adults, and promote peer connectedness. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 22 AL reducing risk ofsuicide. Potential Outcomes Reductions in maladaptive coping attitudes and behaviors Increases in healthy coping attitudes and behaviors lncreaseg in referrals for youth in distressed Evidence Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. Peer norm bpproache Evaluations show that programs such as Sources of Strength can- 5 Community engagement activities. Community engagement is an aspect of sooal capital Community engagement approaches may involve residents participating in a range of activities, including religious activities community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall phySical health, reduced stress, and decreased depressive thereby lncreas Increases in pPositive perception; of adult support help?seeking behaviors Comment IAI: From JH: Cite findings from Let?s Connect intervention program? While the intervention didn?t signi?cantly reduce suicidal behavior, there were signi?cant improvements in connectedness (a protective factor for suicide]- is a great thought. We have quite a few programs that didn?t reduce suicidal behavior, but do impact related RFs, so we?re trying to stray from them and focus on those that impacted suicide and attempts or have a greater amount of >evidencefmore evaluations behind them. Comment From T5: Is there any evidence on outcomes more proximal to suicide? Did they look for any and not find them? It is important to describe relevant null effects when they were found. KH: They did not look at more proximal outcomes, but 505 is currently undergoing a 6-year evaluation funded by NIMH to examine impact, hopefully with suicidal behaviors being one of the houtcomes they? re measuring. Improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted With 18 high schools (6 metropolitan, 12 rural} Wyman et al. (2010] found that the program improved .I adaptive norms regarding suicidegmonggeer leaders, theteconnectedness to adults and school engagement Peer leaders were also more likely than controls to refer a suicidal friend to an adult students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those With a history of suicidal ideation, and the acceptability of help? seeking behaviors- Iseel?i-hg?help. PeceeptierFi?ef these improvements in factors that protect against suicidal behavior translate into reduced suicidal behavior and suicide. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4 436 lots [or 7 8 million square feet} in 4 areas ofthe city. Researchers found significant associated reductions in community residents? self?reported level of stress which is a risk factor 23 for and engagement in more physical exercise, a protective factor for suicide, han residents in control vacant lot areas. Other bene?ts included reductions in firearm assaults and vandalism EBranas et al., 2011} Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem?solving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors {World Health Organization, 2014}. Suicide prevention programs thatfocus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to [either]direct learning, modeling, andggenvironmental and individual hopelessness} characteristics. The literature linking life skills and suicide is robust. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters [Pollock Williams, 2004}. Treatments that include bolstering skills (Goldsmith, Pellmar, Kleinman, St Bunney, 2002iand include problem?solving techniques {Ghahramanlou-Holloway, Bhar, Brown, Olsen, Beck, 2012; Townsend et al., 2001) appear to redUce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents Comment IAI: From JM: Not seeing strong evidence here. Are you suggesting that this affects documented risk factors for suicide by reducing stress and increasing physical exercise? iwould like to see a stronger evidence statement here. From LLD: you might want to incorporate a phrase in the statement along the lines of "which are risk factors for suicide" KH:Added the qualifier you suggested, kLinda. 4.. Comment From T5: They had a follow Up paper in 2013 that showed intervention sites felt significantly safer. The effects on crimes was encouraging but not significant. it seems important to mention this too. From LLD: not sure i agree with Tom?s last point about mentioning the nonsignificant effect on crime. "l Comment From LLD: the word ?either" here implies another comparative clause; should the word ?or" be inserted before the word ?modeling"? May want to just delete the word "either? important life skills to offset the underlying vulnerabilities that contribute to engaging in high-risk behaviors early in life. Approach es Current evidence provides support for the following two approaches: Social emotional learning programs focus on developing and strengthening communication and problemesolving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 8t Webster-Stratton, 2011i. 24 0 Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and familv relationship programs have been shown in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including ones closelv related to suicide, such as depression, internalizing behaviors, and substance abuse (ii/1.5. Knox, Burkhart, El. Hunter, 2010}. Potential Outcomes Reductions in suicide attempts and suicide ideation Improvements in of risk and protective factors associated with suicide Reductions in suicide risk behaviors depression, anxietv, conduct problems, substance abuse} Improvements in?anel noenaa?Iiee?help-seeking behavior ImprovementsEnhaec?e? social competence and emotional regulation skills problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to reduce suicidal behaviors and associated risk factors. Comment [Ah lthink this should be this reference 2015: Comment IAI: From JH: Suggest translating OR and 95% CI for wider audience. wider audience. See Tom's suggestion below. HRevised accordinglv Comment From .IH: state if the control group received anv kind of intervention \Added Comment From JH: Report ages i {compare to 505 below} 1 ?1 From LLD: agree; this is too technical for a Jin. Social emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers that uses interactive dialogue and role-plaving to teach adolescents about the risk and protective factors associated with suicide [including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems. The program includes 3 hours of role-plav sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program randomized controlled trial of YAM conducted across 10 European Union countries and 168 schools, Estudentiaged 14-15 participating in the YAM program were significantlv less likely to?haveani-neideet-suieideatte . . - . attempt suicide and ii ?have severe suicidal ideation . the 12-month follow-up compared it to the kernel youth suicide attempts among the YAM group was reduced bv over 50%; indieat?i-ngdemonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionallv, related to severe suicide ideation, in the 25 v? YAM group absolute risk fell relative risk fell by 49.6% Wasserman et al., From JH: signi?cant? Signs ofSufcr'de {505} is another school-based prevention program for students aged 13-11 The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as well as improve and normalize help- seeking behavior (Schilling, Aseltine, 81 James, 2016). in a randomized controlled trial, $05 was shown to suicide attempts at 3-months post intervention among participating 1 students' knowledge of how to get help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward helprseeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to Eower?risk participants}5chi ling et al., 2016}. 11 It i\ 1 i Finally, the Good Behavior Game is a classroom?based program for elementary school children aged The program uses a team?based behavior management strategy that promotes goodjl?x,? behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior {Wilcox et al., 2008). Two cohorts of youths participated in the program in 1985?35 and 1986- 86 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, ?-aagn outcome evaluation of the GBGJ?indicated that assignedindividuals in the first cohort who were assigned to participate in when they were in the first grade reported half the adjusted odds of suicidal ideation and suicide later, between thejges Esttemptd when assessed approximately 15 years compared to peers who had been in a standard classroom setting. The beneficial effect ofthe program was consistent for suicidal ideation regardless of whether baseline covariates were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. in the?iecond students, neither suicidal idiotion nor suicide attempts were significantly different between 636 and the control interventions {Wilcox et al., 2008]: {FaisThe authors surmise this finding likely arose due to the lack of fl . implementation fidelitqand pointed to the need for 686 to be delivered with precision, y" consistency, and teacher support. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide {Kellam et al., 2008}. 25 I. I I i (Comment From JH: signi?cant? 1 .i - kRevisedComment lAl: From JH: signi?cant? Comment IAI: From T5: The Lancet paper has a 2015 publication date. The authors provide the absolute and RH for suicide attempts too and they give a clear way ofthinking about this- for RR "Of 100D pupils, 11 attempted suicide in the control group vs five attempts in It might be worth using the attempt example so you can provide this explanation. JL. Comment From LLD: Jeff had Comment [Ah From .IH: is this the jffect size/magnitude of reductioniI Comment From JH: define control condition 3 Comment From JH: Lower-risk? Do you mean students with no prior reported history of suicide attempts? questions about what is meant by upstream. You are referring to primary prevention but some ofthe other Comment From JH: compared to? gridded. Comment Hi: From JH: Suggest stating clearly there are two separate cohorts of students receiving the intervention. How do these cohorts differ (time, location}? kn:- .1 Comment [Ah From TS: it is important to help the reader to understand that this study looked at the suicidal ideation when the ?rst graders were age 1941. This is .H authors? comment or your interpretation of the data? Please clarify. 'nnlu I Comment IAI: From signi?cant? Also please de?ne externalizing I Parenting skill and familvr relationship programs. Parenting and family skills training ibehaviors approaches have shown promising impacts in preventing key risk factors associated with 5 i kH Added Comment From T5: This transition suicide. For example, the incredibie Years is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth bv improving protective factors such as responsive and positive parent-teacher?child interactions and relationships, emotion self~regulation and social competence (all protective factors for suicide] (Herman et al., 2011}. The program includes 9? 20 sessions offered in communitv~based settings religious, recreation centers, mental health treatment centers, and hospitals}. Several studies have demonstrated i Seems off. i i the effect of the 0? program on reducing internalizing such as anxietv and i i i i i 3 EH: edited. depression, and child conduct problems (C. H. Webster-Stratton, Reid, 8i. Beauchaine, 2011; Webster?Stratton, Jamila Reid, 8: Stoolmiller, 2008}. The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year followvup [Reid, 8L Hammond, 2003,- C. Hammond, 199?; C. Webster-Stratton, Reid, 8i. Hammond, 2001). The program demonstrated greater benefits as the dosage ofthe intervention increased (Herman et al., 2011}. Additionally, Strengthening Families 1014 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing i emotions and conflict, and communicating with their children; promoting vouths? 'i interpersonal and problem-solving skills; and creating familv activities to build cohesion and i? positive parent?child interactions. The premise of the program is that developing these skills I for both parents and children will reduce internalizing behavior and adolescent substance ii abuse, two important risk factors for suicide (Spoth, Guvll, 8f. Dav, 2002}. Strengthening alcohol use, and drug use among participating families (Spoth et al., 2002). 2? Identify and Support People AtsRisk Rationale In order to decrease suicidegttention to people at increased example people with prior suicide attempts, is necessary, as these individuals tend to experience suicidal behavior at higher than I I average rateslThese vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio-economic status-of-who are living with a mental health problem; people who have attempted suicide previouslv; individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain racial and ethnic minoritv groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective wavs of identifving at-risk or vulnerable groups, customizing services to make them accessible and engaged in care remain kev challenges. For example, simplv improving services does not guarantee that those services will be used bv those most in need of them, nor will it necessarilv increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that mav adverselv affect their abilitv to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifving and supporting people at increased risk. Gatekeeper training is designed to train teachers, coaches, providers and others in the communitv to identifv people who mav be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is tvpicallv implemented in schools to identify at-risk youth and within health care settings to identifvf adults (and youth}. Screening combined with care management and overall continuitv of care has been used in primary care and behavioral health care settings to assure that people who may be at high?risk of suicide are identified and receive ongoing treatment as needed, particularlv after inpatient discharge and other transitions within the healthcare svstem so thev don?t 'slip through the cracks?. These approaches tvpicallv emplov screening for depression andlor suicide combined with collaborative treatment planning between patients and their providers and patient follow? up. Crisis intervention. These approaches provide support and referral services, tvpicallv bv connecting a person in crisis [or a friend or familv member of someone at?risk} to trained Volunteers or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact kev risk factors for suicide, including feelings of 28 Comment IAI: From TS: This seems i r' 1. awkward as written. it might be better to say to people with speci?c Comment Alz i?rn not sure about ?specific vulnerabili?ties? terminology. I think people are much more familiar with something like this. depression, hopelessness, and subsequent mental health care utilization. Like means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Potential Outcomes Reductions in suicide attempts Reductions: in suicide deaths in identification of individuals at?risk for suicidal behavior teereasedincreases in at-risk individuals in treatment i-neseased?lncreases in community members trained to identify at?risk individuals teeseased?Increases in referrals for health care Evidence Identifying people at risk of suicide can positively impact both suicide risk factors as well as suicide mortality, however sustained implementation of programs and practices appears critical. Among people who id entity themselves through calls to suicide hotlines, preliminary evidence finds reductions in suicide sweides-eu-t?thee?eets?mest?be u- - - wu- suicidal behavior is needed [Kitchener 8: Jorm, 2006}. II Gatekeeper training. One example of gatekeeper training is the Mental Heaith First Aid program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises andior in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders], possible risk factors, and where and how to getEasdelenee?based ?ll effectivelhelp [Kitchener 3i Jorm, 2004}. In a randomized controlled trial of 301 participants of {ii MHFA, the. ii'? stars; [137.53% in}; 336?]? E?i'?'rii?iif? ?113 ?33? follow?up, significantly greater feelings of con?dence gc??ll, greater likelihood of encouraging people to seek professional help 29.4% vs. 16.8% i improved agreement with health professionals about treatments [9&036), and 5 decreased stigmatizing attitudes towards mental illness ip<.02l. Additionally, the intervention 1 Comment From LLD: suggest 5 shortening this introductory statement 'l about the evidence and making it i consistent with the other sections. if you need to provide caveats, then you might want to take a look at the TP for Lexample wording. Comment From TS: Add cites here. i Also this wording is awkward because it could be read as suggesting that those who chose to use the hotline are different from others at risk who did not. Maybe you could add the follow up period to the end -they have decrease suicidal Lthoughts and behavior at Kit {awkward here i possible to report magnitude of if "f intervention effects? I JL 1 Comment IAE: From LLD: wording is a bit Comment From .IH: signi?cant? Is it Comment From JH: Overall mental etc]? Please clarify .1 I i ?if percent who provided some or a lot of help did not differ between groups (Kitchener 81 Jorm, 2004]. Additional research rigorously evaluating MHFA for its impact on intervention recipients? 29 I I health or specific facets depression. Comment IAI: From JM: I think you need to provide a little more context on i Henry Ford so the readers understands in . what kind of population the reductions were experienced. I assume from the end . that Henry Ford is some sort of HMO, but just a little more information would help clarify this description to readers. r[Comment [Ah From JH: how much? Gatekeeper training has also been a primary part of all Garret Lee Smith (GL5) Suicide Prevention Program which is in place in 49 states and 43 tribes. i5. multi?site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation [Walrath, Garraza, Field, Goldston, SI McKeon, 2015]. This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237I deaths in the age groUp, between 2007 and 2010. Counties implementing GL5 program activities, including gatekeeper training, also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GLS activities {4.9 fewer attempts per 1000 youths; [Godoy Garraza, Walrath, Goldston, Reid, 8: McKeon, 2015)}. More than 29,000 suicide attempts may have been prevented during the period examined, following implementation of the GL5 program. I Screening combined with care management and overall continuity of care. [The Henry Ford healthcare system is a large health maintenance organization (HMO) in the state of Michigan. Hang: Ford?s Perfect Depression Care program was the pre-cursor to what is now called Zero Suicide, and its overall goal was to eliminate suicide among its members. More broadly, the aim was to redesign delivery of depression care to achieve ?breakthrough improvement" in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide (C. E. Coffey, 2006]. An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82%]50 E. Coffey, 2006; C. E. Coffey, Coffey, 8i. Ahmedani, 2013). Further?: among HMO members who received mental health specialty services, the suicide rate decreased from 110.3/100.000 in 1999 to 47.6f100,000 in 2010 with a mean of 362,000,000 over i the time period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased, from12/100,000 to 5.6f100,000 [p<.01l. Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 981100.000 to 12.5f100,000 [p<.001i (M. Coffey, Coffey, 8f. Ahmedani, 2015). 30 It Crisis intervention. Suicide prevention hotlines are one wav to provide crisis intervention. In an {Comment From JH: Report evaluation of the effectiveness of the National Suicide Prevention Lifeline to prevent [i magnitude 0* decrease? suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment l: Comment This is bestl Fl can do. Not sure if vou want this level of detailhovvever. for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days . (mean=13.5 davs] after the initial assessment. Researchers found that over half of the initial i; Comment To be consistent i above I added some measures of {r sample were seriouslv considering suicide when they called, and they had a plan fortheir suicide. Researchers also found that among follow?up participants, there was a signi?cant hecreasdinj pain p<.001l, hopelessness lF=47.8, p<.001l and intent to die lF:7.57, ps?ll between initiation of the call {time 1} to follow?up {time 3). Between time 2 lend of the to time 3, the effect remained for pain lF=14.1, and hopelessness association. I I I paDDll but was not significant for intent to die. (Gould, Kalafat, Harrismunfakh, 8: Kleinman, 2007]. i i in another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, i Munfakh, and Kleinman (2013) assessed the impact of the Applied Suicide intervention Skills Training a widelvr implemented training program that helps hotline counselors, i emergency' workers, and other gatekeepers to identifv and connect with suicidal individuals, i understand their reasoning for living and diving, and assist with safelv connecting those in need i to available resources. The training was evaluated across the NSPL network of hotlines over the i period 2008-2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care ithout training in were significantly more training, individuals who spoke with counselors likelv to feel depressed suicidai [1.39, more overwhelmed {1.18, 1.32}, and leg hopeful [1.351134 1.7?ll by the end of their call to the hotline compared to those with training in A5!5T.k3ounselors trained in were also: more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. 31 intervene to Lesseu Harms and Prevent Future Risk Rationale Comment IAI: From LLD: simplify statement and make it consistent with the other introductory statements about approaches. You may also want to take a look at the introductory statements to this section in the other TF5. J's. 5 Individuals who have experienced mental heaith challenges, suicidai ideation, who have made suicide i attempts or engaged in non-suicidal self-injury are at increased risk of suicide (US. Office of the Surgeon i General 84 National Action Alliance for Suicide Prevention, 2012]. Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide lPitman, Osborn, King, St Erlangsen, 2014). Exposure to sensationalited or uninformed reporting regarding on i: suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute i to suicide contagion (Etzersdorfer 8: Sonneck, 1998; Niederkrotenthaler El Sonneck, 200?]. i i i Approaches EQ?beead-areay?ef-aThe four approaches included here to lesson harms and reduce future risk of suicide among those at increased risk focus on providing appropriate mental healthcare, continuity of care, caring for the bereaved, and providing safe messaging around suicide. I may want to mention how these are typically delivered leg, one on one or group formats; typical number of sessions, location, etc}. omment [Ah From Suggest eshing this description out a bit. You Included some ofthis info. Idon'tthink there's a "typical" number of sessions, as it seems to vary across treatment . . .- .. -. - a .--.-. n- - I- - - reatment for people at?risk of suicide ]typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem~solving, impulsivity and emotion regulation. Treatment usualiy takes place in a one on one or group format between patients and clinicians in mental healthcare settings and can vary in duration from several weeks to ongoing therapy for years in some cases. Treatment to prevent These approaches typically include follow-up contact and use diverse modalities leg, home visits, mail, teiephone, e?mail] to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on?one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare,- 12%?25% reattempt within a year, and 396?996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015] - Postvention approaches are implemented after a suicide has taken place and may include debrie?ng sessions, counseling, and/or bereavement support groups for surviving friends and 32 ?settings. Comment IAI: From add something about contagion? family members/loved ones. These programs have not typically been evaluated for their impact KH: added on suicide or suicidal behavior but may reduce survivors' guilt, feelings of depression, and complicated grief (Saumilas 8i. Kutcher, 2011). Safe messaging following a suicide. The manner in which information on a recent suicide is i i communicated to the public leg. school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide and suicide contagion. Potential Reductions in mental health-related sequrelae I Increasesiij connectedness Improvements in coping skills Improvements in messaging following suicide I Reductions in re-attempts I Reductions in contagion effects related to suicide Evidence The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide?related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended followrup, much ofthe evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of impact on risk and protective factors for suicide. One example is the improving Mood? Promoting Access to Coiioborotive Treatment program. JMPACT aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. it facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et al.. 2006}. The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2006; Unutzer et al., 2006) relative to patienE who received care as usual. 33 Another example is Coiiaborative Assessment andr Management ofSuicio?oiity which is a . lei: From 1-5: lthinka word is therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This ,1 "lizltl??'f flexible approach can be used across treatment settings and clinician theoretical orientations and Added. involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative andiinvoive constant if patient input about what is and is not working with the ultimate goal of enhancing the i" i' a therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 5 correlational studies (lobes, 2012), in a varietv of inpatient and outpatient settings and in one RCT with several additional under wav. CAMS has been associated with significant improVEments in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow?up among a communitvbased sample of suicidal outpatients. (Comtois et al., 2011). Other examples include Diaiecticai Behaviorai' Ti?ierapjvr (DST) and Attachment-Based Familyr Therapy (ABFT). a multicomponent therapv for individuals at high risk for suicide and who may struggle with impulsivitv and emotional regulation. The components of include individual therapv, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DST were half as likely to make a suicide attempt at two?veer follow-up than women receiving communiti,r treatment [23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006). a program for adolescents aged 12-18 and is designed to treat clinicallv diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxietv {Diamond et al., 2010]. A randomized controlled trial of ABFT found that suicidal adoIEscents assigned to ABFT experienced significantiv greater improvement in suicidal ideation over 24 weeks of follow?up than did adolescents assigned to enhanced usual care. Additionally, a higher percentage participants reported no suicidal ideatlon in the week prior to assessment at 12 weeks than did adolescents receiving enhanced Usual care [69.2% vs. 34.6%} and at 24 weeks (82.1% vs. 46.2%] (Diamond et al., 2010]. Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief intervention with Foliow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined with nine follow-up contacts over 13 months (at 1, 2, 4, 7, 11 weeks and 4, E, 12, 18 months]. Follow?up contacts are either conducted bv phone or through home visits according to a Specific time line for up to 13-months. A randomized 34 controlled trial that enrolled suicide attempters from eight hospital emergency departments in I, [five culturally different countries [BraziL India, 5ri Lanka, Iran, and China]sites]found that a brief intervention combined with 9 follow?up visits over Iii?months was fewer deaths from suicide relative to a treatment-as-usual group versus respectivelvl (Fleischmann et al., 2008]. Another example of treatment to prevent re~attempts involves octive foiiow?up contact approaches such as postcards, letters, and telephone calls intended to increase a patient's sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typicallv invite patients to reconnect with their provider. Contacts are made periodically or everv few months in the first 12 months post? discharge with some programs continuing contact for two or more veers]. In a meta-analvsis conducted by Inagaki et al. {2015], interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by i approximatelv 17% for up to 12 months post?discharge; however, the these; approaches over periods of time longer than one vear on reattempts has not 1vet been demonstrated. Also. because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow- up approaches on death bv suicide. In a randomized controlled trial of the post?crisis suicide ?d Comment IAI: From .IH: Unclear. to the article, this program was evaluated in five countries. Please state the countries. Tho I. Jidded. Comment From JH: De?ne [follow- _i ups greater than 1 veer] I kRevised. Comment IAI: From JH: De?ne "1 In. KH: Defined later in the sentence kcontact 4var, 2 vaars of followI up. prevention follow-up contact approach, Motto and Bostrom (2001} found that patients who refused ongoing care but who were randomized to be contacted bv letter four times per veer had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post?crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam, Sarjami, Kolahi, Carter, 2011; Wang et al., 2015}. Finally, Cognitive Behavior Theropyfor Suicide Prevention is an example of a therapeutic approach to prevent re-attempts. It Uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related dif?culties] leading up to and following the suicidal event; safety plan development; skill building; and also has familyr skill modules focused on familv s. A support and communication patterns as well as improving the familv?s problem solving skill randomized controlled trial of found that ill?session outpatient cognitive the designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood suicide reattempt among adults who had been admitted to an emergency' department suicide attempt relative to treatment as usual [Brown et al., 2005I- 1 Postvention programs are implemented with the goal of providing support to survivors of others' Comment w: From Don?t you mean "Recommendations for Reporting on Suicide"? Please cite: suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service l5tondBy}, provides clients with face-torface outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Uisser, lComans, Scuffham, 2014). In a study by Visser et al. [2014), StondBy clients were i significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group i who had not had contact with the Stond?y program (48% and 64% respectively}. Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive _i postvention [versus passive approaches where survivors self-refer for services] ll. Cerel 8i i Campbell, 2008}. Safe messaging following a suicide. Safe messaging after a suicide can help assure that reporting of the event is clone in such a way to reduce risk to consumers of news media and other messaging who may be particularly vulnerable. lElne way to ensure safe messaging following a suicide is to encourage that reporters adhere to Recommendations for ?eporting on inner Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources leg, hotline) and that avoid sensationalizing events or reducing suicide to one cause Ir Comment From T5: Are you sure The media did not want "guidelines" and they reacted negatively to the idea of external guidelines. KH: Edited accordingly. I didn?t realize they didn?t like the idea of "guidelines" - kinteresting. Comment itbink this is a better link.) 1 I that the 81 reduction wasjust on the subway system? lthought that was nationally. Please confirm. KH: it was nationally apologies for the koversight in the way this was written. Comment IAI: From T5: Thomas also had a 2010 report that suggested benefits of good reporting. KH: Added. can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of recommendations for reporting on suicide on reduction in suicides comes from Austria. After a sharp increase in suicides on the subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides pnnualltl (Niederkrotenthaler 8i Sonnet: way leg, reporting on suicide and repetition} have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects Reports of individual suicidal ideation not against suicide [Niederkrotenthaler et al., 2010]. accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, associated with significant decreases in suicide rates in the time period immediately following such reports {Niederkrotenthaler et al. 2010]. 35 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal], social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, health care insurers and providers, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Access to Mental Health Cure. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non?governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been at the forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting 37 Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Probiem-Soiving Skiils to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support Peopie At?Risir and to Lessen Harms and Prevent Future Risk. The intensity and activities of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health care, social services, and justice sectors can work collaboratively to support individuals at high- risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 38 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation as planning, implementation, and assessment all rely on accurate measurement of the problem. Surveillance data helps researchers and practitioners track changes in the burden ofsuicide. Surveillance systems exist at the federal, state, and local levels. it is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certi?cates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches [Blair, Fowler, Jack, Crosby, 2016}. The National Electronic Injury Surveillance System-All Injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc), age, racefethnicity, sex, disposition (where the injured person goes when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health?risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013}. The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides Mnational and state?level estimates of substance use, including alcohol, tobacco, illicit drugs, and non-medical use of prescription d?gg; and mental health, including past year mental illness, Ito-occurring illnesses, service utilization, along with suicide ideation suicide plans, and suicide pttempts} NSDUH resource tohtrack trends}! in suicide?related risk factors in the population and to help identify groups at increased risk. It is also important at all levels (local, state, and federal] to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk Comment From TS: This description seems incomplete comparted to the description of YRBS ?1 Comment IAI: From TS: Should this be and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. programs, and '50 be consistent? '3 The evidence-base for suicide prevention has advanced greatly over the last few decades. However, r: {comment FromJH;including additional research is needed to understand the impact of Earogram olicies and practices on Euicide ?amp?? 1 'l Comment [Ali .IH: redundant With I prior sentence. {and suicide attempts, at a minimum}, 5 opposed to merely examining thei_r effectiveness risk factors I: asseeiated?wish?sui-eide. More research is also needed to examine the effectiveness of upstream before risk occurs! and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive Prevention approachConclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far?reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis-at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. Unfortunately and unlike most other public health problems, suicide mating Mstill overcome sesaggiesagaiastm stigma, shame, and secrecy related to help-seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear efrglatgd_t9_asking someone about their risk of suicide [versus the fear and consequence of not asking), and?fear of taking up strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, misplaCed emphasis by the media and others on sensational aspECts of suicide, and disproportionate ,t_s public health burde?Wbeng??gW funding given effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive fashionComment From TS: lfound this I somewhat awkward. The idea of suicide struggling seemed off and the point is not entirely clear. I think you can be more specific and reference an earlier point Comment From .IH: i is jargonComment prefer to keep this language in if it?s ok. The e.g. is meant to help define the terms and it's been mentioned previously and defined. It's also a continued topic of discussion in the field that we need to focus more upstream. Its-rating and re-iterating it is good I think. And Iastiy, there?s a lot of jargon and terms in here that we'd also i I ill I I Lthe media recommendations section. have to get rid of if we?re honest about Lthe use ofjargon and reading levelcombination-4n a multi-level, multi-sectoral way. This technical package includes strategies and- I approaches targeting preventiorilegu social emotional learning for children and youth), as. well as strategies focused more cognitive behavioral treatment to prevent re- 40 .J attempts}. It includes that range from a focus on the whole population regardless of risk to strategies designed lmportantly, this technical package extends the bounds of the typical prevention strategies to consider approaches go beyond individual behavior change to better address risk factors impacting communities and populations more broadly housing and community engagement initiatives. While the evidence base continues to emerge, the collection of programs, policies, and practices laid outlifxn, here are available for implementation now. And in keeping with good public health practice, the intent l?i is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state ofthe science. In closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? Comment From TS: Ithink you can drop this and just use the more Lcouid describe this as a range. Comment Hi: From JMLYES this is jargon, so the helps, but you might want to figure out a way to say this without referring to the outer levels of ,__the social ecology. Comment This sets you up for people to say oh yeah what transgendered people, prisoners etc. I. Comment lAl: l'm not exactly clear on how to take this comment and subsequently how to address it. Open to 41 Lideas. descriptive test. You don't need both. You 1 JL References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 42 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2015d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (201Ee). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel, J., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, J., Maple, M, DE VENNE, J. V., Moore, M., Flaherty, C., 31 Brown, M. {2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Health Reports, 131(1). Cerel, J., McIntosh, J. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of "Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and late-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. J., Borges, G. L. 6., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research, 28(5 SUPPL.), 185-285. Classen, T. J., 8: Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass~layoffs and unemployment duration. Health Econ, 21(3), 338650. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524?1526. Coffey, C. E., Coffey, M. J., 8; Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. doi:10.1176iappi.PS.640422 Coffey, M, Coffey, C., 8: Ahmedani, B. K. (2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963?972. doi:10.1002fda.20895 Coulton, C. J., Crampton, D. 5., Irwin, M., Spilsbury, J. C., 8: Korbin, J. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Child Abuse Neal, 11111142. Cox, (3. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M, . . . Pirkis, J. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. Crosby, A. E., Espitia-Hardeman, V., Ortega, L., Loza no, El. (2013). Alcohol and suicide. Alcohol: Science, Policy and Public Health, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., 8: Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., at Melanson, C. (2011}. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. (2002). Exposure to Suicide: Incidence and Association with Suicidal ldeation and Behavior: United States, 1994. Suicide and life-threatening behavior, 32(3), 321-328. Cunningham, P. J. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Health Aff {Millwood}, 28(3), w490-501. 43 Curtin, 5. C., Warner, M., at Hedegaard, H. (2016}. increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., St Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am Epidemioi, 180(1), 45-52. 106 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zvvi, 84 R. Lozano Worid Report on Vioience and Heaith (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, 5., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of vioience, 140. Deisenhammer, E. A., lng, C. M., Strauss, R., Kemmler, (3., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am} 152(2), 311-318. Diamond, (5. 5., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, 5. (2010). Attachment?based family therapy for adolescents with suicidal Ideation: a randomized controlled trial. Am Acad Chiid Adaiesc 49(2), 122131. Dunne, E. J., McIntosh, J. L., 8: DunnevMaxim, K. (Eds). (1987'). Suicide andits aftermath: Understanding and counseiing the survivors. New York: Norton. Durkheim, E. [1897i1951]. Suicide: A study in sociology (JA Spaulding at G. Simpson, trans). Giencoe, it: Free Press.(0riginai work pabiishea? 189?). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 8; Vella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adaiesc Heaith, 45(4), 351?359. Escobedo, L. (3., 81 Drtiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anai Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1993). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. 0., Hindman J., Reed, G. A., Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 81 Radcliff, B. (2009). Public policies and suicide rates in the American states. Sociai indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buiietin of the Worid Heaith Organization, 86(9), ?1034709. doi:50042A 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., 8; Zhou, C. (2015). Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortaiity weekiy report, 64(38), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 8i Frazier, L. (2015}. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Pubiic Heaith, 105(2), 311-316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Chiid Moitreot, 11(3), 263-280. doi:10.117711077559506289524 44 Ghahramanlou-Holloway, M., Bhar, S., Brown, 6., Olsen, C., Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-302. Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., 81. McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Goldsmith, 5. K., Pellmar, T. C., Kleinman, A. M., Bunney, W. E. (2002). Reducing suicide: A nationai imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Ha rruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411?421. Han, E., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 8 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205228. Harris, E. C., 8; Barraclough, B. (1998). Excess mortality of mental disorder. 173, 11?53. Harris, K. M., Carpenter, C., 8 330, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. HassanianeMoghaddam, H., Sarjami, S., Kolahi, A. A., Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Journai of 198(4), 3094316. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi:101027702226910.28.51.4 Heise, L., Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano Worid Report on Vioience and Heaith (pp. 82-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q, 26(3), 189-201. Hong, J. S., Kral, M. J., Espelage, D. L., St Allen-Mea res, P. (2012). The social ecology of adolescent-initiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431-454. doi:10.1007f510578- 011-0273-y Houle, J. N., Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am i Puhiic Heaith, 104(6), 1073-1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-253. 45 lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. JAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Isometsa, E. T. (2001). autopsy studies--a review. Eur 16(7), 379-335. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Clin Exp Res, 33(12], 2124-2133. Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, 5. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohai Depend, 95 Suppi 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. G., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engli Med, 352(24}, 2515-2523. Kitchener, B. A., 8.1mm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 3.1mm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aast ZJ 40(1), 6?8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 8: Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273?229. 5910/a000136 Klevens, J., Simon, T. R., Chen, J. (2012]. Are the perpetrators of violence one and the same? Exploring the co occurrence of perpetration of physical aggression in the United States. interpers Vioience, 27(10), 198742002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ, 32717423), 1326. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Pubiic Heaith, 100(12), 245742463. Knox, M. 5., Burkha rt, K., 84 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment?related parenting behaviors and beliefs. Journal' of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Heaith Econ, 22(1), 23-88. doi:10.1002ihec.1816 Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. Losel, F., St Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. 5., Cluispe-Agnoli, M., Ouyang, L., 8; Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am .i Pabiic Health, 101(6), 1139-1146. 46 Maimon, D., Browning, C. 8., 8; Brooks-Gunn, l. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Shern, D. L., Bagalman, J. E., Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, lv?A: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8c Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Marla Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. {2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. G., 8: Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Motto, J. A., 8: Bostrom, A. G. (2001]. A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., St Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: 0.5. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., 8L Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21 41. cloi:10.1080f00048670701266680 Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. Olfson, M., Gerhard, T., Huang, C., 5., 8t Stroup, T. 5. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of selfeharm: Systematic review. British Journal of 131(3), 193?199. Perron, 5., Burrows, 5., Fournier, M., Perron, P. A., St Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 Pinchevskv, G. M., 8t Wright, E. M. (2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma 1Liialence Abuse, 13(2), 112-132. Pirkis, 1., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 8r. Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. 8., 8: Williams, J. M. {2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. 47 Reeves, A., Stuckler, D., McKee, M., Gunnell, 0., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. 1., Webster-Stratton, C., 8: Hammond, M. (2003}. Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. 0., Ireland, M., 8-1 Borowskv, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Studv of Adolescent Health. JAdolesc Health, 35(5), 424 e421? 410. Rowhani-Rahba r, A., Simonetti, J. A., 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. doi:10.1093fepirevimxv006 Ru nvan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidalitv. Westi Emerg Med, 17(1), 8-14. Runvan, D., Wattam, C., Ikeda, R., Hassan, F., 8: Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Loza no (Eds), World Report on Violence and Health (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. 0., 8i. Gannon-Rowlev, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review ofSociology, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 84 James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 152?166. Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, G. A., Jr., 81 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352?362. doi:10.1111i5ltb.12225 Silverman, M. ML, 8L Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and life? threatening behavior, 25(1), 10?2 1. Simon, 0. R., Swann, A. (2., Powell, K. E., Potter, L. E., Kresnow, M. J., 3: O'Carroll, P. W. (2001). Characteristics of impulsive Suicide attempts and attempters. Suicide Life Threat Behov, 32(1 Suppl), 49?59. Spoth, R. L., Guvll, M, 3L Dav, S. X. (2002). Universal family?focused interventions in alcohol~use disorder prevention: costneffectiveness and costnbenefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 228. Stack, S., Wasserman, I. Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 37(1), 103?112. Sudak, H., Maxim, K., Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. Szumilas, M., Kutcher, S. (2011). Post~suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem-solving treatments after deliberate self-harm: meta-a nalvsis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979-988. Tvrer, P., Reed, G. M., 8: Crawford, M. J. {2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Office of the Surgeon General, 84 National Action Alliance for Suicide Prevention. (2012). 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. 48 Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 42(1), 19-28. doi:10.1002fjcop.21586 Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016). Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 105-120. Wasserman, Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 1536-1544. Webster-Stratton, C., Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison ofchild and pa rent training interventions. Consult Clin 55(1), 93-109. Webster-Stratton, (3., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283-302. Webster-Stratton, C. H., Reid, M. J., 8: Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child 8: Adolescent 40(2), 191*203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008}. Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high?risk schools. Journal of child and 49(5), 471?488. Widome, H., Sieving, R. E., Harpin, S. lit, at Hearst, M. O. (2008). Measuring neighborhood connection and the association with violence in young adolescents. JAdaiesc Health, 43(5), 482?439. Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, l. M., Ialongo, N. 5., Wang, W., &Anthony,1. C. (2008). The impact of two universal randomized first? and second~grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, $6043. Wilson, W. J. (2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A, Brown, C. H., LoMurray, M., Schmeelkaone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653?1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, Wu, K., 8L Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 49 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Other Risk] Protective Factors for Suicide Suicide Attempts or ldeation Lead Sectors1 Government (local, Unemployment benefit programs state, Federal) Strengthen economic Other income supports Businessflabor . . . . . . Housmg pollcles Government {locaL The National Neighborhood Stabilization state, Federal) Program Coverage of mental health conditions in health insurance policies Health ca re Strengthen Mental Health Parity Laws 1/ access to Government (state, mental health Fadera? ca re Reducing access to lethal means among persons at-risk Government (local, intervening at suicide hot spots state) 1/ Safe storage practices Public Health Establish protective Organlzatlonal pollcles and culture Busmesstabor . Together for Life envnronments Government (local, US Air Force Suicide Prevention Program 1/ state, Fed eral) 50 Best Available Evidence Communitywbased policies to reduce excessive alcohol use Alcohol outlet density 1/ Government {local, state] Businessllabor Promote connectedness Peer norm approaches Public Health Sources of Strength Education Communitvrengagement activities Public Health Greening vacant urban spaces Government {local} Social emotional learning Public Health Youth Aware of Mental Health Program Education Signs of Suicide Teach co in 3 Good Behavior Game and problem- solvmg Sklils Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening Families 10-14 Gatekeeper training Public Health Mental Health First Aid Healthcare Screening combined with care management Healthcare Identify and 51 support people alt-risk Henry Ford Perfect Depression Core {Pre- cursor to Zero Suicide) Best Available Evidence Social Services Crisis Intervention Public Health Active foiio w?up con toct approaches EST for Suicide Prevention Notionoi Suicide Prevention Lifeiine Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide Health ca re improving Mood Promoting Access to v? Social Services Coiioborotive Treatment Justice Coiiohorotive Assessment and Management v? Suic?doii CAMS Intervene to I lessen harms Dioiecticoi Behovioroi Therapy and prevent future risk Attachment?Based Fomiiv Therapy Treatment to prevent re-attempts Health ca re ED Brief intervention with Foiiow-up Visits Social Services v? Postvention Heaithcare 52 Best Available Evidence StandBy Response Service Safe messaging following a suicide Media Guidelines 1/ Public Health Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 Policies, Programs, and Practices to Support individuals, Families, Communities: A Technical Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technicai Package to Prevent Suicide. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Strengthen Access to Mental Health Care 16 Create Protective Environments 18 Promote Connectedness 22 Teach Coping and Problem-Solving Skills 24 Identity and Support People At-Risk 28 Intervene to Lessen Harms and Prevent Future Risk 32 Sector Involvement 37 Monitoring and Evaluation 39 Conclusion 40 References 42 Appendix A 50 Acknowledgments [to be inserted later] External Reviewers [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called seif?directeo? vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, 81 Melanson, 2011). Self- directed violence may be suicide! or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fatal self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. in 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016d). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U5 (Centers for Disease Control and Prevention, 2016d). Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, 8; Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by raceiethnicity with the highest rates, across the lifespan, occurring among non-Hispanic American lndiaanlaska Native (rate: Among young people are disproportionately at increased risk of suicide with young males aged 25-29 experiencing the highest rates (rate: (Centers for Disease Control and Prevention, 2016dl. Moreover, suicide rates among non?Hispanic have increased by 48.7% since 1999. Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, 8t Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press}. Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological mode ?- encompassing multiple levels of focus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002}. Risk and protective factors for suicide exist at each level. For example, risk factors include: individual level: History of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization?d perpetration, and genetic and biological determinants Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General at National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). It is important to recognize that the vast majority of individuals who are depressed or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes (Butchart, Phinney, Check, 8; Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide (Desai, Dausey, 8L Rosenheck, 2005) as well as perpetration of child maltreatment (Coulton, Crampton, Irwin, Spilsbury, 8i. Korbin, 2007; Freisthler, Merritt, 8: LaScala, 2006), teen dating violence (Capaldi, Knoble, Shortt, 8: Kim, 2012), intimate partner violence (Pinchevsky 8: Wright, 2012), and youth violence (Sampson, Morenoff, 8: Gannon-Rowley, 2002). Additionally, a lack of economic opportunities and unemployment are associated with suicide (Luo, Florence, Quispe-Agnoli, Ouyang, 8-: Crosby, 2011; Reeves et al., 2012), as well as perpetration of child maltreatment (D. Runyan, Wattam, lkeda, Hassan, 81 Ramiro, 2002), intimate partner violence (Heise 8: Garcia-Moreno, 2002; Pinchevsky Wright, 2012), sexual violence (Centers for Disease Control and Prevention, 2016c) and youth violence (Wilson, 2011). Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills (Centers for Disease Control and Prevention, 2016a, 2016c, 2016e; U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012). Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, 8 hmma? connectedness 1? W?s Borowsky, Hogan, 8: Ireland, 1997; Centers for Disease Control and Prevention, 2016b; Coulton et al., 2007; Kleiman, Riskind, Schaefer, Weingarden, 2012,- Pinchevsky 84 Wright, 2012; Widome, Sieving, Harpin, 8t Hearst, 2003), school (Basile, Espelage, Rivers, McMahon, 8t Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Williams, 2007,- DeGue et al., 2013; Hong, Kral, Espelage, &Allen?Meares, 2012; Losel 8-1 Fa rrington, 2012), family [Ca paldi et al., 2012; Centers for Disease Control and Prevention, 2016a; Elgar, Craig, Boyce, Morgan, St VellavZarb, 2009; Maimon, Browning, 8i Brooks-Gunn, 2010; Resnick, Ireland, 8L Borowsky, 2004], caring adults (Capaldi et al., 2012; Losel 8: Farrington, 2012; Maimon et al., 2010}, and pro?social peers (Capaldi et al., 2012,- Losel 3i. Farrington, 2012i enhances resilience to Suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far-reaching consequences for individuals, families, and communities (Dunne, 8t Dunne- Maxim, 198?; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015]. Man early estimated that 7% of the US. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that [gigachsgic?gil? adults were exposeigmnew about the dgth. in a more recent study, in one bysuicide in their lifetimes. Research indicates that the impact of who died bysuicide andfo_r having lived experience he. having attempted suicide oneself} iimuch more extensive than injury and death. Suicide attempt survivors may suffer long?term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt [Chapman St Dixon?Gordon, 200?}. Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, Mortimer- Stephens, 2004}, stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide [Julie Cerel, McIntosh, Neimeyer, Maple, St Marshall, 2014; Sudak, Maxim, 8: Carpenter, physical tolL survivors of suicide financially and occupationally (Florence, Simon, Haegerich, Luo, 8: Zhou, 2015], The economic toll of suicide is immense as well. According to more conservative estimates, in 2013, suicide cost 150.3 billion in estimated lifetime medical and work?loss costs [florence eLal., 2015]. By another estimate the total lifetime costs associated with nonfatal injuries and deaths caused by self- directed violence in 2013 were approximately $93.5 billion after adjusting for under?reporting of suicide 5 {Shepard, Gurewich, Lwin, Reed, Silverman, 2015)] The overwhelming burden of these costs results: from lost productivity over the life course, with {Shepard et al., 2016). .- Cerel et. al [2015] found that 48% of the weighted survey population knew at least one person who died 5 [Comment IAI: From T5: I found this wording somewhat awkward. Comment From TS: it seems like it would be appropriate to acknowledge the CDC estimate too. Consider using that as a minimum and then say that others have estimated that the costs could be considerably higher after ad lusting for under-reporting ofsuicide. Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable (US. Public Health Service, 1999]. And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that suicide will not be prevented by any single intervention taking place in any single setting {Silverman St Maris, 1995; U.5. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, labor) {National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014]. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a] meta-analyses or systematic reviews showing impact on suicide; b} evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; meta?analyses or systematic reviews showing impact on risk or protective factors for suicide, or d] evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms ofthe nature and quality of the available Comment IAI: From TS: ?developmental nature"I doesn't seem correct. Do you mean the ?current statusevidence. Not all programs, policies, or practices that utilize the same approach ie.g., gatekeeper training} are equally effective, and even those that are effective may not work across all populations. 10 Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. In practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security 0 Housing stabilization policies Strengthen access to mental health care Coverage of mental health conditions in health insurance policies Create protective environments 0 Reducing access to lethal means among persons at-risk of suicide Iv Organizational policies and culture I Community?based policies to reduce excessive alcohol use Promote connectedness . Peer norm approaches I Community engagement activities Teach coping and problem-solving skills Ir Social-emotional learning I Parenting skill and family relationship approaches Identify and support people at risk - Gatekeeper training . Screening combined with care management Crisis intervention Treatment for people at-risk of suicide Treatment to prevent re-attempts Postvention intervene to lessen harms and prevent future risk I I II 11 0 Safe messaging following a suicide The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support may also vary by population and community characteristics. Ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator le.g., strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 1 Comment IAI: From T5: was thinking ,3 that it might be good to work in a point about how these risks can be directly or indirectly associated with suicide risk because these stressors can exacerbate relationship as well physical and mental health problems. [Comment From JH: Consistent tense reduced vs. reduce Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- I age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015). Economic and ?nancial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and want to consider using similar language housing expenses, and even the anticipation of such financial stress, can directly or indirectl? increase an individual?s risk for suicide; buffering these risks can therefore, potentially protect against suicide :5 for your potential outcomes. {Stack 8: Wasserman, 200?}. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and ll medical care, job training, child care, among other expenses required for daily living. In providing this ll support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing :3 Comment From LLD: the other packages use ?RedIJctions you may I I suicide. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household ?nancial security can potentially bufferthe risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples ofways to strengthen household financial security. a Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential LutcomeJ - Reduced suicide rates a Lower foreclosure rates Lower eviction rates Reduced emotional distress 13 Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. An examination of variations in US. unemployment bene?t programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits {Cylus, Glymour, Et Avendano, 2014]. Another U.S. study examining the linlt between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed tojust the loss of job, predicted suicide risk {Classen 3d. Dunn, 2012]. Together, these results suggest that not only should state unemployment benefit programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance] have also shown an impact on suicide. A study by Flavin and Radcliff [2009) examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance {Temporary Assistance to Needy Families and total state spending on suicide rates between 19902000, controlling for a number of suicide risk factors leg, residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn't spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin 8t Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased their per ca pita spending on these types of assistance $45 per yearKFiavin 8t Radcliff, 200931 Housing stabilization policies. The National Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration of the neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working?aged adults [Houle 8t Light, Another study 14 Comment IAI: From TS: Consider including a point about this being a correlational study and more evaluation Work is needed but it suggests the potential benefits of policies that reach those Iwho are particularly Vulnerable at the times when they are in greatest need. of data from 16 US. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began) to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 1' Comment From .IM: I assume this means that different types of mental Strengthen Access to Mental Health Care I: health problems pose different risks for suicide, but wasn't sure? May want to Lmake the point more explicit JL Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, 3 Comment From T5: Isn't this including substance abuse disorder? it Huang, Stroup, 2015; Owens, 2002}, had risk conferred by mental illnesses ,5 Lapierre, Kim, 31 Turecki, 200d; E. C. Harris 8: Barraclough, 199?; Tyrer, Reed, St Crawford, 2015), previous if Lwould be good to state this explicitly. research indicates that mental Illness is an Important risk factor for suicide (E. C. Harris Barraclotugh, Comment Hi: From TS: I found this 1998} World Health organization, 2014). that Up to 90% 0f people Who die by SUICICIE 1" I sentence confusing Can you reward ta may have had a mental ?llnesd at the time of their deaths {ArsenauIt-Lapierre et al., 2004,- Cavanagh, ,l From LLD: could say: "if a state has a Weaker parity law than the federal parity law leg, includes coverage for some mental health conditions but not others}. then the federal parity law will replace Lthe state law.? Carson, Sharpe, 81 Lawrie, 2003; Isometsa, 2001}. State?level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shem, Bagaiman, 8: Cap, 200?]. Findings from the National Comorbidity Survey indicate that relatively few people in the US. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one ofthe contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention {World Health Organization, 2014}. Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns Ii.e., mental health parity}. Benefits and services covered include such things as the number of visits, co?pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. [Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law ie.g., includes coverage for some mental health 5 conditions but insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes Increased utilization of mental health services 15 II Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. Decreased of mental illnesses Decreased rates of suicide attempts Decreased rates of suicide a Comment IAI: From TS: i don?t think this is accurate. I was curious about this study and checked out the paper. I think they are saying that this is the cost incurred per suicide prevented. They go on to explain that this is the upper bound and does not reflect the benefits to non? suicidal individuals. Please con?rm this. If you are going to include this then it will Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Sao (2006) found that 12 months after states enacted mental health parity iaws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013} examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year. Lang et. al (2013} estimated the cost of saving a life through such mandates as mental health parity by comparingthe loss inwages attributable to the policy, via increased premiums, to the number of lives saved. Based on these life was between?and Feast?savingsvef?lj?SJ million?L However, this calculation did not take into account the saving associated with improved mental increased mental health care. As such, this figure is considered the upper bounds on costs incurred and should be interpreted with cautionpeHuieide?peevented (Lang, 2013]. Lbe important to explain it further. Create Protective Environments Rationale Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddom 1980}. Creating environments that address risk and protective factors where individuals live, work, and plav, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2016; Mcintosh et al., 2016), therefore, changes to the organizational culture in these occupations, bv wav of implementation of supportive policies or evan phvsical modifications to the workplace environment, can change social norms, encourage help?seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not L. Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarlv, modifving the characteristics of the environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis (Beautrais, Gibb, Fergusson, Horwood, 31. Larkin, 2009,- A.E. Crosby, Espitia~ Hardeman, Ortega, Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenwav, 8t Azrael, 2015; C. W. Runvan et al., 2016}. Approaches The current evidence suggests three bromisin?approaches for creating environments that protect; against suicide. Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumping from heights provide little opportunitv for rescue and, as such, have high case fatalitv rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that lithe interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001} and 2) that people tend not to substitute a different method when a highlv lethal method is unavailable or difficult to access lHawton, 200?; Yip et al., 2012]. Therefore, increasing the time interval between the thought and the suicide attempt, for example, bv Comment IAI: From lid: The word promising has certain connotations when using it around evidence. lr?ou sav later that the evidence around this strateg'v is some of the strongest for suicide. These two statements seem somewhat inconsistent. i would use a different word here then promising lf vou truly believe there is strong evidence. From JH: suggest deleting the word ?promising? From LLD: since we don't want to infer that these are the onlyr approaches for creating protective environments, I might be inclined to change the word ?promising" to ?potential? or "poSsible" i I i tor something along those lines. making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening otSuicio?e Hotspots. Suicide hotspots. or places where suicides may take place relativelv easilv, include tall structures leg, bridges and cliffs), railwav tracks, and 18 isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help (Cox et al., 2013}. Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable andfor impulsive from easy access to lethal means. Such practices may include edUCation and counseling around storing firearms?docked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8t Rivara, 2016; C. W. Runyan et al., E3- 2015). Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior le.g., asking for help], skill building, changing social norms, referral and access to helping services mental health, substance abUse treatment, financial counseling}, and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) {National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol?involved suicides (Escobedo 8t Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, takes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one?third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004]. Potential Outcomes Increase in safe storage of means Reduction in suicide attempts Reduction in suicide deaths Increase in help?seeking Reduction in alcohol?related suicide deaths Evidence The evidence for the effectiveness reducing access to lethal means for person at-risk of suicide and other ways to establish protective environments Es some ofthe strongestlin the field Zalsman etal.,20161Comment IAI: From .Il'v1: Consistent with promising used earlier? Is it the strongest? Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Oueliet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseiing on Access to Lethoi Means ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre?post quality improvement project, Runyan et al (2016} found that at post-test (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara St Martin, 2012). Police suicides were tracked over 12 years and 20 compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 28.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 100,000) (Mishara 8: Martin, 2012). Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement}. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, Caine, 2003). Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2003 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community?based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8: Remer, 2009). 21 Comment IAI: From JM: Seems like a nuance you can delete from this paragraph. You don?t really pick up on this later and it probably requires more Promote Connected ness Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897;1951]. Connectedness is the degree to which an ,i individual or group of individuals are socially close, interrelated, or share resources with others (Centers ocial connections can be formed within and between for Disease Control and Prevention, 2009). multiple Ieveis ofthe social ecology (Dahlberg 0 Krug, 2002}, for instance between individuals leg. peers, neighbors, co-workers], families, schools, neighborhoods, workpiece, faith communities, cultural groups, arid mile? as a community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8L Laud, 2015; Muennig, Cohen, Palmer, 81 Zhu, 2013). Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. ENhile the ?ha_ Comment Ml: FromJH; Awkward 1 explanation wording. Consider, while the evidence is Comment From JM:ThlsimpIies that prosocial behavior is currently i abnormal. Maybe promote is a better word? From JH: suggest deleting ?seek to normalize? and say "Peer norm approaches encourage prosocial Comment IN: I prefer to keep it as normalize and took out prosocial behavior. We want to normalize help- limlted, existing studies 4k. seeking versus making it seem like a evidence is still being built]me pattern is towards a positive ninveeseassociation between social capital measured by social Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole {Centers for Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. u?naa? Peer norm approaches seek lo factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. These approaches typically target youth and are delivered in school settings but can also be impIEmented in community settings. Lpersonal weakness. I I I I 22 I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean?up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby Comment IAI: From JH: Cite findings from Let's Connect intervention program? While the intervention didn?t signi?cantly Comment From TS: Is there any reduce suicidal behavior, there 'were signi?cant improvements in connectedness (a protective factor for JR. suicide], evidence on outcomes more proximal to suicide? Did they look for any and not ?nd them? It is important to describe relevant reducing risk of suicide. Potential I - Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed i a Increase help?seeking behaviors ii . . . i: - Posmve perception of adult support li i ii . . i bwdence 5: I I I Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. 0 Peer norm Evaluations show that programs such as Sources of Strength cani i Reduction in maladaptive coping attitudes and behaviors Outcomes improve school norms and beliefs about suicide that are created and disseminated by student Lnull effects when they were found. ., Comment IAI: From JM: Not seeing strong evidence here. Are you suggesting that this affects documented risk factors for suicide by reducing stress and increasing physical exercise? I would like to see a stronger evidence statement here. From LLD: you might want to incorporate a phrase in the statement along the lines Lof "which are risk factors for suicide? Comment From T5:They had a i follow Up paper in 2013 that showed i intervention sites felt significantly safer. The effects on crimes was encouraging but not significant. It seems important to mention this too. From LLD: not sure I agree with Tom's last point about mentioning the nonsigniflcant peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high?schools (E metropolitan, 12 rural}, Wyman et al. [2010] found that the program improved peer leaders' i adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer i i leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, i the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained et al., 20105. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,435 lots [or?.3 million square feet} in 4 areas ofthe city. Researchers found significant associated E'eductions in community residents' self?reported stress levels and engagement in more physical exercise han residents in control vacant lot areas. Other benefits included I reductions in firearm et al., 2011} 23 Leffect on crime. ;,{Comntent Speci?c edits needed I here. Comment From LLD: the word Teach Coping and Problem-Solving Skills Rationale . Building life skills prepares individuals to successfullv tackle everv dav challenges and adapt to stress and "either? here implies another - - - - - - I comparative clause,- should the word "or" I . I advers tv Life skills encompasses manv concepts, but most often include coping and problem solving If he Inserted before the "modeling?? skills conflict resolution, and critical thinking. Life skills are important in shielding individuals from ill" May want tolust deletethe word "either? suicidal behaviors (World Health Organization, 2014}. Suicide prevention programs thatfocus on life and fl social skills training are drawn from social cognitive theories (Bandura,1985},surmising thatEuiciclaLf,Ir behavior is attributed to Either ]direct learning, modeling, and environmental and individual leg; hopelessness} characteristics. The literature linking life skills and suicide is r.obust The tnabilitv to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters [Pollock EL Williams, 2004). Treatments that include bolstering skills (Goldsmith, Pellmar, lKlelnman, Eh Elunnev, 2002}and include problemvsolving techniques {Ghahramanlou-Hollowav, Bhar, Brown, Olsen, Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at-risk populations. While manv do not target suicidal behaviors directlv, these programs strive to train vouth and parents in important life skills to offset the underlving vulnerabilities that contribute to engaging in high-risk behaviors earlv in life. Approaches Current evidence provides support for the following two approaches: Social emotional learning programs focus on developing and strengthening communication and problem?solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. Thev provide children and vouth with skills to resolve problems in relationships, school, and with peers, and help vouth to address other negative influences substance use} associated with suicide. These approaches are tvpicallv delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 8L Webster-Stratton, 2011}. I Parenting skill and familvr relationship programs are designed to strengthen parenting skills, enhance positive parent?child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and familv relationship programs have been shoWn in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 24 ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8t Hunter, 2010}. Potential Outcomes Reduction in suicide attempts and suicide ideation Enhanced knowledge of risk and protective factors associated with suicide Reduction in suicide risk behaviors li.e., depression, anxiety, conduct problems, substance abuse) Improve and normalize help-seeking behavior I I Enhance social competence and emotional regulation skills - Enhance problem-solving and conflict management skills Evidence There are several programs with evidence that support teaching social, emotional and parenting skills to [Comment IAI: From .IH: Report ages {compare to 505 below) r. Comment From JH: Suggest Ll translating DR and 95% CI for wider if audiEnce, From LLD: agree; this is too technical for a wider audience. See Tom's suggestion ll ll :1 if Lbelow. Comment From JH: state if the ll control group received any kind of ll fLintervention If I lli Comment From JH: signi?cant? Comment [Ah From Lancet 'l paper has a 2015 publication date. The authors provide the absolute and RR for suicide attempts too and they give a in the control group vs five attempts in RR ?Of 1000 pupils, 11 attempted suicide in. 2 reduce suicidal behaviors and associated risk factors. program developed for teenagers that uses interactive dialogue and role-playing to teach Social emotional learning programs. The Youth Aware of Mental Health Program is a i; i 5 clear way ofthinking about this for i i it might be worth using the attempt example so you can provide this adolescents about the risk and protective factors associated with suicide {including i knowledge about depression and anxiety} and enhances their problem-solving skills for Lexplanation, dealing with adverse life events, stress, school and other problems. The program includes 3 ill {Comment Hi: From JH: signi?cant? ii: i Effect sizelmagnitude of reduction? hours of role-play sessions and interactive workshops combined with a booklet that students can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program [Wasserman et al., 2014). in a cluster- schools, Etudentiparticipating in the TAM program were significantly less likely to have anIi, I incident suicide attempt EUR 0.45, 0.24?0.85; p=0.014) and severe suicidal ideation i :l {0.50, 0.27-0.92; p=0.025l ]at the 12-month follow-up compared to the ontrol grou Additionally, related to End relative risk fell by?g?irs EWasserman et al., 2014} :l Signs ofSuicide {505) is another school-based prevention program for students aged 13-17. The program includes guided classroom discussions about suicide and depression. As part of ll the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as Well as improve and normalize help- seeking behavior (Schilling, Aseltine, St James, 2016). in a randomized controlled trial, 505 was shown to suicide attempts at 3-months post intervention among 25 I I l' I I participating students compared to [:ontrol students. The SOS program also increase students' knowledge of how to get help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help?seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to ?ower-risk participants]5chi ling et al., 2016}. Finally, the Good Behavior Game {636) is a classroom?based program for elementary school i .4 .I [Comment From JH: signi?cant? 0 r? ,4 children aged 6-10; it represents an example of upstream suicide prevention programming The program uses a team?based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the (336 is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior {Wilcox et al., 2003). in an outcome evaluation of the 636, first graders assigned to 636 reported half the adjusted odds of suicidal ideation and suicide [attemptd The beneficial Jr effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 686 effect on attempts was less robust in some adjusted models including caregiver mental health. in the hecond cohort 10f 686 students, neither ,r'l 3" suicidal ideation nor suicide attempts were significantly different between 636 and the i l' I control interventions (Wilcox et al., 2008} ifhis finding likely arose clue to the lack of implementation fidelity] and pointed to the need for (33-5 to be delivered with precision, consistency, and teacher support. (336 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide {Kellam et al., 2008}. Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent?teacher?child interactions and relationships, emotion self?regulation and social competence {all protective factors for suicide] (Herman et al., 2011]. The program includes 9? 20 sessions offered in community?based settings religious, recreation centers, mental health treatment centers, and hospitals}. Several studies have demonstrated the effect of the li? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, 8t Beauchaine, 2011; Jamila Reid, 8: Stoolmiller, Elli-03). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year followrup [Reid, Webster?Stratton, 8; Hammond, 2003; C. 8; HammondComment IAI: From define control condition Comment [Ah From JH: Lower-risk? Do you mean students with no prior reported Ihistory of suicide attempts? Comment From LLD: Jeff had questions about what is meant by upstream. You are referring to primary prevention but some of the other examples YAM are also primary prevention. Suggest just deleting this Lstatement. tommentji?: From JH: compared Comment IAI: From .IH: Suggest stating clearly there are two separate cohorts of students receiving the intervention. How Ldo these cohorts differ (time, location}? Comment From TS: it is important to help the reader to understand that this study looked at the suicidal ideation when the ?rst graders were age 19-21. This is m_otclear currently. Comment From is this the authors? comment or your interpretation Lof the data? Please clarify. I 1997; C. Webster?Stratton, Reid, Sn Hammond, 2001). The program demonstrated greater . gamma?. lAl: benefits as the dosage of the intervention increased (Herman et al., 2011). 3 N50 de?ne externalizing behaviors. i Additionally, Strengthening Families Iii-l4 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem?solving skills; and creating familv activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guvll, 8i. Day, 2002]. Strengthening Families has been shown to ecrease externalizing hehaviord1 alcohol use, and drug use if among vouth participants gangreductions in depression, participating families [Spoth 2? Identify and Support People AtoRisk Rationale populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously; individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain racial and ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at?risk or vulnerable groups, customizing services to make them accessible and engaged in care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults {and youth}. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t ?slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. - Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reductionorder to decrease suicide, Lttention to people at increased is necessary as these individuals tend to experience suicidal behavior at higher than average rates.]These vulnerable or disadvantaged 5:3" Fa Comment IAI: From TS: This seems awkward as written. it might be better to say to people with speci?c Comment IAI: lprefer to use the standard language here. Comment IAI: From LLD: suggest shortening this introductory statement about the evidence and making it consistent with the other sections. If you need to provide caveats, then vou might want to take a look at the TP for Lexample wording. Comment From TS: Add cites here. Also this wording is awkward because it could be read as suggesting that those who chose to use the hotline are different crisis interventions can put space or time between an individual who may be considering suicide JL . I and harmful behaworPotential Outcomes Reduction in suicide attempts [Comment [Ah From JH: compared to? - Reduction in suicide deaths I Increased identification of individuals at-risk for suicidal behavior 0 Increased at?risk individuals in treatment 5 . . I Ir from others at risk who did not. Maybe 0 Increased communltv members trained to identify at-risk Individuals l5 You mum add the In?ow up period to the II Increased referrals for health care I I end -thev have decrease suicidal Lthoughts and behavior at KK -: 1 l5 Comment IAI: From LLD: wording is a bit LI awkward here ii I Comment From .IH: signi?cant? Is it possible to report magnitude of intervention effects? 1 1 55: Comment IAI: From JH: Overall mental health or specific facets depression. etc)? Please clarify I I Evidence . - ggv-ln a- - I Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid program. This program is designed for the lav public and consists of three weeklv sessions of three hours each. Participants learn the of people in mental health crises andfor in the earlv stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, xietv, and disorders), possible risk factors, and where and how to get effective an Fwdenee?based?e?eedvsi?hel?p?p (Kitchener 81 Jorm, 2004). In a randomized controlled trial of 3019 participants of intervention group, compared to the wait?listed controls, }n providing help to others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about 1 Jorm, 2004}. Additional research rigorously evaluating MHFA for its impact on intervention recipients? suicidal behavior is needed {Kitchener 3L Jorm, ZDUEII Gatekeeper training has also been a core part of all Garret Lee Smith Suicide Prevention Program which is in place in 49 states and 43 tribes. A multi-site evaluation assessed the 29 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among the population aged 10?24 in counties implementing GLS trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties implementing GL5 trainings had significantly lower youth suicide rates the year following the training implementation (Walrath, Garraza, Reid, Goldston, 8a. McKeon, 2015]. This finding represents a decrease of suicide death per 100,000 among youths 10 to 24 years of age or the avoidance of approximately 23? deaths in this age group between 2007' and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youths 15 to 23 years of age in the year following implementation of the GL5 program than did similar counties that did not implement GLS program activities (4.9 fewer attempts per 1000 youths; {Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015)). More than 79,000 suicide attempts may have been averted during the period studied following implementation of the 1315 program. I Screening combined with care management and overall continuity of care. [The Henry Ford Perfect Depression Care program Was the pre-cursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program detreloped concrete measures to assess progress on each of these aims and began with screening and assessment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the Henry Ford Perfect Depression Care {Pm-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention [1999 and 2000] to the intervention years (2002-2009). During this time period, the suicide rate fell E. Coffey, 2006; C. E. Coffey, Coffey, 8L Ahmedani, 2013]. Further, suicide rates also declined aniE?'g'i?ii?o'i?'? and received mental health specialty services. However, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increasecl(M. Coffey, Coffey, 8r. Ahmedani, 2015]. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notional Suicide Prevention Lifeline to preven suicide, 1,035 suicidal individuals who called the hotline completed a standard risk assessmen for suicide, and 330 of those completed a follow?up assessment between 1 and 52 days after the -.J: Comment IAI: From JM: think you need to provide a little more context on I Henry Ford so the readers understands in I: what kind of population the reductions were experienced. I assume from the end that Henry Ford is some sort of HMO, but just a little more information would help clarify this description to readers. [Comment [Ah From JH: how much? {Comment From JH: Report magnitude of decreaseinitial assessment. Researchers found that over half of the initial sample were seriouslylI considering suicide when they called, and they had a plan for their suicide. Researchers also; found that participants experienced significant Elia-creased in suicidality over the course of the} 30 1 telephone session, and that levels of hopelessness and pain continued to decrease Eafter their initial Kalafat, Harrismunfakh, 8; Kleinman, 200?]. ln another study, this time employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, and Kleinman (2013) assessed the impact of the Appiied Suicide intervention Training (MIST), a widelyr implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify.r and connect with suicidal individuals, understand their reasoning for living and dving, and assist with safelv connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in ASIST were signi?cantlv more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful bv the end of their call to the hotline. Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013]. 31 ,"(Comment From .IH: how long? 1 Comment From LLD: simplify statement and make it consistent with the other introductory statements about approaches. ll'ou may also want to take a look at the introductory statements to ,this section in the other TF5, Intervene to Lessen Harms and Prevent Future Risk Jk Comment Ml: From LLD: Suggest ?ashing this description out a bit. You may want to mention how these are typically delivered one on one or group formats; typical number of Lsessions, location, etc]. Rationale Individuals who have experienced mental health challenges, suicidal ideation, who have made suicide attempts or engaged in non-suicidal self-injury are at increased risk of suicide (US, Office of the Surgeon General 84 National Action Alliance for Suicide Prevention, 2012]. Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide lPitman, Osborn, King, St Erlangsen, 2014]. Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion lEtzersdorfer E: Sonneck, 1998; Niederkrotenthaler Si Sonneck, 200?]. Approaches Pt broad array of approaches to lesson harms and reduce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow~up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing communications to emphasize resilience, decrease negative affect, and to prevent contagion] ll In [Treatment for people at-risk of suicide ltypically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. I Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities leg, home visits, mail, telephone, e?mail] to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on?one interpersonal therapy andy?or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296?2596 reattempt within a year, and Bit?99$ of attempt survivors die by suicide within 1 to 5 years of their initial attempt (inagaki et al., 2015] I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors' guilt, feelings of depression, and complicated grief (Stumilas 8t Kutcher, 2011]. 32 Potential buxom-as} {Comment IAI: From .IHL add something about contagion? Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide i f: contagion. Therefore, responsible and safe reporting may help prevent suicide and suicide contagion. Reduction in mental health?related sequelae Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts Evidenco The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow?up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of impact on risk and protective factors for suicide. One example is the Improving Mood- Promoting Access to Coiiaborative Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. It facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et al., The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideatian over 24~months of follow-up {Hunkeler et al., Unutzer et al., 2006?; relative to patients who received care as usual. Another example is Collaborative Assessment and Management which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 33 batientlabout what is and is not working with the ultimate goal of enhancing the therapeutic H, alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported' In 6 correlational studies (lobes, 2012), In a variety of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community?based sample of suiCIdal outpatients. (Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DIST) and Attachmenb?ased Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include Individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two year follow up than women receiving community treatment [23% vs redoired less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006). ABl?Tis a program for adolescents aged 12?18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety {Diamond et al, 2010] A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage ofABFi" participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care [69.2% vs. 34.6%} and at 24 weeks (82.1% vs. 45.2%] (Diamond et al., 2010]. I Treatment to prevent re-attempts. Several strategies that aim to prevent re?attempts have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief lnterveation with Follow?up ILlisits is a program that involves a one?hour discharge Information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior, and referral options, combined Iwith nine follow?up contacts over 13 months (at 1,2,4, 11 weeks and 4, E, 12, 18 months]. Follow- up contacts are either conducted by, - phone or through home visits according to a specific time line for up to 18? months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in 13 months was associated with significantly fewer deaths from suicide relatIve to a treatment- as?usual group versus respectively] {Fleischmann et al., 2008Comment IAI: From TS: thinka word is missing? "patient input"? Comment IAI: From JH: Unclear. According to the article, this program was evaluated in five Countries. Please state the countries. Thmt. Another example of treatment to prevent re-attempts involves active follow-up contact 1 approaches such as postcards, letters, and telephone calls intended to increase a patient's sense 1" of connectedness with health care providers and decrease isolation. These approaches include [Comment De?nE expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically leg, or every few months in the first 12 months post? discharge with some programs continuing contact for two or more years}. In a meta-analysis conducted by Inagaki et al. {2015]. interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the [ong?term effectskifthese approaches on reattempts has not yet been demonstrated. Also, because the number of trials 1? and associated sample sizes included in this meta~analysis were small, it was not possible to determine the effect of active contact and follow?up approaches on death by suicide. in a 1i: randomized controlled trial of the post?crisis suicide prevention approach, Motto and Bostrom {2001} found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-lvloghaddam, Sarjami, Kolahi, Carter, 2311; Wang et al., {Comment IAI: From .IH: De?ne {follow- ups greater than 1 year] 2016]. Finally, Cognitive Behavior TherapyforSuicide Prevention is an example ofa therapeutic approach to prevent re-attempts. It uses a riskvreduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties] leading up to and following the suicidal event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial of found that lU-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% redUction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual [Brown et al., 2005]. I Postvention programs are implemented with the goal of providing support to survivors of others? suicide to reduce their own risk of suicide. One example of a postvention program, StandBy Response Service [StondBy], provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, E: Scuffham, 2014). In a study by Visser et al. [2014}, StandBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group 35 1 who had not had contact with the Stand?y program (43% and 64% respectivelv]. Additionallv, ,FCommem From 1-5: gonrww mean research suggests that active postvention approaches in which outreach to suicide survivors "Recommendationsfor?eporting occurs at the scene of a is associated With Intake Into treatment sooner, greater Pissing i afs .or attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors selfsrefer for services] (J. Cerel E: i i ,i The media did not want ?guidelines" and they reacted negativelyI to the idea of Lexternal guidelines. Comment From TS: Are Ivou sure that the 31 reduction Wasjust on the subway system? i thought that was knationallv. Plea5e confirm. Comment IM: From JH: other evaluations available? . Comment IAI: From TS:Thomas also Campbell, 2003}. Safe messaging following a suicide. Safe messaging after a suicide can help assure that reporting of 4 the event is done in such a wav to reduce risk to consumers of news media and other messaging who may,f be particularly vulnerable. One way to ensure safe messaging following a suicide is to encourage that reporters adhere to media guideiines for reporting on Euicided, Re ports that are If both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline) and that avoid sensationalizing events or 'r had a 2010 report that suggested bene?ts reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guideiines on reduction in suicides comes 5. if! ofgood reporting. 3? b' .rc ch.or content b' rc I I from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the 19? ,3 23?me df guidelines on subsequent suicides. Changes in the clualiti;r and quantity of media reporting resulted in a significant reduction of 81 suicides pnnuallquin-thejfiennese sot-arrayI pvstemi' JL KNiederkrotenthaler 8i. Sonneck, 20073 35 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, health care insurers and providers, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Access to Mental Health Care. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 37 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People At?Risk and to Lessen Harms and Prevent Future Risk. The intensity and activities of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health care, social services, and justice sectors can work collaboratively to support individuals at high- risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 38 Comment From This description seems incomplete comparted to the Monitoring and Evaluation i description of YRBS Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation as planning, implementation, and assessment all rely on accurate measurement of the problem. i Surveillance data helps researchers and practitioners track changes in the burden ofsuicide. Surveillance systems exist at the federal, state, and local levels. it is important to assess the availability of surveillance i data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National 'v'iolent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certi?cates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, 3: Crosby, 2016}. The National Electronic Injury Surveillance System-All Injury Program provides 5 nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc), age, racefethnicity, sex, disposition (where the injured person goes i i i i I when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health?risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013}. The mass data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Health is an annual nationwide survey of individuals aged 12 years and older that provides national and state?level estimates of drug use and mental health-related issues, including suicide ideation and suicide Ettempt? It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long? term outcomes are an important part of program evaluation. 39 ?i The evidencevbase for suicide prevention has advanced greatly over the last few decades. However, Comment lei: From Ts:should this be programs, policies, and practices to be consistent? 1 additional research is needed to understand the impact of programg policies, and practices on .1 [Comment IAI: From JH: including preventing [suicide and suicide attempts,]as opposed to merely examining I to impact risk factors associated with suicide. More research is also needed to examine the effectiveness attempts? of upstream and community?level strategies to prevent suicide at the population level. Lastly, it will be 1 Comment From .IH: redundant with rior sentence. important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but if Comment lAi= From 15: Hound this if somewhat awkward. The idea of suicide ,7 struggling seemed off and the point is not there is potential to understand the synergistic effects within a comprehensive prevention approach. strategies? and If entirely clear, Ithink you can be more i i specific and reference an earlier point in Lthe media recommendations section. ., Comment [Ah From .IH: i? :l Comment From T5: lthink you can i drop this and just use the more is jargon. 1 I I descriptive text, You don't need both, You Conclusion ,couid describe this as a range. Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, 3 its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, :llilil those with lived experience, and those with thoughts of suicide, on a daily basis??at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done ii. to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name ?l just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles .l against stigma, shame, and secrecy related to help~seeking, mental illness, being a survivor, or someone i includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination--in a multi-level, multi-sectoral way. This technical package includes strategies and if? l0I-. ion Erin ?i:r .h?d In: :U'i well as strategies focused more cognitive behavioral treatment to prevent re- attemptsi. It includes [.iniversal, selective, and indicated strategies, orktrategies that focus on the whole 5 population regardless of risk to strategies that focus on those grows at highest risk. Importantly, this tEChnical package extends the bounds ofthe typical prevention strategies to consider approaches at the dd Esoter levels of the social ecology} policies to stabilize housing and communitv engagement From JMWES thisis initiatives. - jargon, so the helps, but vou might want to ?gure out a wav to sav this While the evidence base continues to emerge, the collection of programs, policies, and practices laid out r'e?cer'rlng the outer ?f . . . x? DCIEI eco ogv. here are available for Implementation now. And in keeping With good public health practice, the intent \r Comment This sets you up for is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence people to m, oh yeah what becomes available, this technical package can be refined to reflect the current state ofthe science. WAN. transgendered PEOPIE: Prisoners Letc. In closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 41 References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 42 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2015d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (201Ee). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel, J., 8: Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, J., Maple, M, DE VENNE, J. V., Moore, M., Flaherty, C., 31 Brown, M. {2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Health Reports, 131(1). Cerel, J., McIntosh, J. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of "Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and late-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. J., Borges, G. L. 6., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research, 28(5 SUPPL.), 185-285. Classen, T. J., 8: Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass~layoffs and unemployment duration. Health Econ, 21(3), 338650. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524?1526. Coffey, C. E., Coffey, M. J., 8; Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. doi:10.1176iappi.PS.640422 Coffey, M, Coffey, C., 8: Ahmedani, B. K. (2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963?972. doi:10.1002fda.20895 Coulton, C. J., Crampton, D. 5., Irwin, M., Spilsbury, J. C., 8: Korbin, J. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Child Abuse Neal, 11111142. Cox, (3. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M, . . . Pirkis, J. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. Crosby, A. E., Espitia-Hardeman, V., Ortega, L., Loza no, El. (2013). Alcohol and suicide. Alcohol: Science, Policy and Public Health, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., 8: Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., at Melanson, C. (2011}. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. (2002). Exposure to Suicide: Incidence and Association with Suicidal ldeation and Behavior: United States, 1994. Suicide and life-threatening behavior, 32(3), 321-328. Cunningham, P. J. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Health Aff {Millwood}, 28(3), w490-501. 43 Curtin, 5. C., Warner, M., at Hedegaard, H. (2016}. increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., St Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am Epidemioi, 180(1), 45-52. 106 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zvvi, 84 R. Lozano Worid Report on Vioience and Heaith (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, 5., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of vioience, 140. Deisenhammer, E. A., lng, C. M., Strauss, R., Kemmler, (3., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am} 152(2), 311-318. Diamond, (5. 5., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, 5. (2010). Attachment?based family therapy for adolescents with suicidal Ideation: a randomized controlled trial. Am Acad Chiid Adaiesc 49(2), 122131. Dunne, E. J., McIntosh, J. L., 8: DunnevMaxim, K. (Eds). (1987'). Suicide andits aftermath: Understanding and counseiing the survivors. New York: Norton. Durkheim, E. [1897i1951]. Suicide: A study in sociology (JA Spaulding at G. Simpson, trans). Giencoe, it: Free Press.(0riginai work pabiishea? 189?). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 8; Vella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adaiesc Heaith, 45(4), 351?359. Escobedo, L. (3., 81 Drtiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anai Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1993). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. 0., Hindman J., Reed, G. A., Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 81 Radcliff, B. (2009). Public policies and suicide rates in the American states. Sociai indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, 0., De Leo, 0., Bolhari, J., Botega, N. J., . . .Thanh, H. T. {2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buiietin of the Worid Heaith Organization, 86(9), ?1034709. doi:50042A 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., 8; Zhou, C. (2015). Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortaiity weekiy report, 64(38), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 8i Frazier, L. (2015}. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Pubiic Heaith, 105(2), 311-316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Chiid Moitreot, 11(3), 263-280. doi:10.117711077559506289524 44 Ghahramanlou-Holloway, M., Bhar, S., Brown, 6., Olsen, C., Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06), 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-302. Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., 81. McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Goldsmith, 5. K., Pellmar, T. C., Kleinman, A. M., Bunney, W. E. (2002). Reducing suicide: A nationai imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Ha rruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411?421. Han, E., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 8 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205228. Harris, E. C., 8; Barraclough, B. (1998). Excess mortality of mental disorder. 173, 11?53. Harris, K. M., Carpenter, C., 8 330, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. HassanianeMoghaddam, H., Sarjami, S., Kolahi, A. A., Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British Journai of 198(4), 3094316. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi:101027702226910.28.51.4 Heise, L., Garcia-Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8: R. Lozano Worid Report on Vioience and Heaith (pp. 82-121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q, 26(3), 189-201. Hong, J. S., Kral, M. J., Espelage, D. L., St Allen-Mea res, P. (2012). The social ecology of adolescent-initiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431-454. doi:10.1007f510578- 011-0273-y Houle, J. N., Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am i Puhiic Heaith, 104(6), 1073-1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-253. 45 lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. JAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Isometsa, E. T. (2001). autopsy studies--a review. Eur 16(7), 379-335. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Clin Exp Res, 33(12], 2124-2133. Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, 5. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Aicohai Depend, 95 Suppi 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. G., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engli Med, 352(24}, 2515-2523. Kitchener, B. A., 8.1mm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 3.1mm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aast ZJ 40(1), 6?8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 8: Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273?229. 5910/a000136 Klevens, J., Simon, T. R., Chen, J. (2012]. Are the perpetrators of violence one and the same? Exploring the co occurrence of perpetration of physical aggression in the United States. interpers Vioience, 27(10), 198742002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ, 32717423), 1326. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Pubiic Heaith, 100(12), 245742463. Knox, M. 5., Burkha rt, K., 84 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment?related parenting behaviors and beliefs. Journal' of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Heaith Econ, 22(1), 23-88. doi:10.1002ihec.1816 Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. Losel, F., St Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. 5., Cluispe-Agnoli, M., Ouyang, L., 8; Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am .i Pabiic Health, 101(6), 1139-1146. 46 Maimon, D., Browning, C. 8., 8; Brooks-Gunn, l. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Shern, D. L., Bagalman, J. E., Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, lv?A: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8c Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Marla Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. in) Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. G., 8: Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Motto, J. A., 8: Bostrom, A. G. (2001]. A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., St Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. (2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: 0.5. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., 8L Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21 41. cloi:10.1080f00048670701266680 Olfson, M., Gerhard, T., Huang, C., S., 81 Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 11724181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193?199. Perron, 5., Burrows, S., Fournier, M., Perron, P. A., 8; Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 Pinchevskv, G. M., 8: Wright, E. M. (2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Violence Abuse, 13(2), 112-132. doi:10.1177l1524838012445641 Pirkis, 1., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 84 Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., at Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Pollock, L. 8., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, 0., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. 47 Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. D., Ireland, M., 8-: Borowsky, I. (2004). ?iouth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdolesc Health, 35(5), 424 e421- 410. Rowhani-Rahba r, A., Simonetti, J. A., 8-: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West.f Emerg Med, 17(1), 8-14. Runyan, D., Wattam, C., Ikeda, R., Hassan, F., 8: Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 8L R. Lozano (Eds), World Report on Violence and Health (pp. 59-86). Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. D., St Gannon-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review ofSociology, 28, 443-473. Schilling, E. A., Aseltine, R. H., Jr., St James, A. (2016). The 505 Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, 6. A., Jr., 84. Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352?362. Silverman, M. M., 84 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and life? threatening behavior, 25(1), 10?21. Simon, 0. R., Swann, A. (2., Powell, K. E., Potter, L. 3., Kresnow, M. J., 3: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M, at Day, 5. X. (2002). Universal familyrfocused interventions in alcohol~use disorder prevention: cost~effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 223. Stack, 5., 8i. Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 37(1), 103-112. Sudak, H., Maxim, K., 8! Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 135-141 doi:10.1176fappi.ap.32.2.136 Szumilas, M, 3; Kutcher, S. (2011). Post?suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, D., Hazell, P., . . .Van Heeringen, K. (2001). The efficacy of problem?solving treatments after deliberate self-harm: meta?analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979?988. Tyrer, P., Reed, G. M., 8: Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Office of the Surgeon General, St National Action Alliance for Suicide Prevention. (2012). 2012 National Strotegyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journalof Community 42(1), 19-28. doi:10.1002fjcop.21586 48 Walrath, C., Ga rraza, L. (3., Reid, H., Goldston, D. B., 8-: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105iAJPH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 1536-1544. Webster-Stratton, C., 8; Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin 55(1), 93-109. Webster-Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 233-302. Webster-Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for 1young children with ADHD. Journal of Clinical Child 8: Adolescent 40(2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schools. Journal of child and 49(5), 471?488. Widome, R., Sieving, R. E., Harpin, S. A., 8: Hearst, M. O. (2008). Measuring neighborhood connection and the association with violence in young adolescents. Adolesc Health, 43(5), 482?489. Wilcox, H. C., Kellam, S. 6., Brown, C. H., Poduska, l. M., Ialongo, N. S., Wang, W., &Anthonv, J. C. (2008). The impact of two universal randomized first? and second?grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?23. Wilson, W. J. (2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A., Brown, C. H., LoMurrav, lvl., Schmeelk?Cone, K., Petrova, M, Yu, (1., . . . Wang, W. (2010). An outcome evaluation ofthe Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653?1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8; Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 49 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Other Risk] Protective Factors for Suicide Suicide Attempts or ldeation Lead Sectors1 Government (local, Unemployment benefit programs state, Federal) Strengthen economic Other income supports Businessflabor . . . . . . Housmg pollcles Government {locaL The National Neighborhood Stabilization state, Federal) Program Coverage of mental health conditions in health insurance policies Health ca re Strengthen Mental Health Parity Laws 1/ access to Government (state, mental health Fadera? ca re Reducing access to lethal means among persons at-risk Government (local, intervening at suicide hot spots state) 1/ Safe storage practices Public Health Establish protective Organlzatlonal pollcles and culture Busmesstabor . Together for Life envnronments Government (local, US Air Force Suicide Prevention Program 1/ state, Fed eral) 50 Best Available Evidence Communitywbased policies to reduce excessive alcohol use Alcohol outlet density 1/ Government {local, state] Businessllabor Promote connectedness Peer norm approaches Public Health Sources of Strength Education Communitvrengagement activities Public Health Greening vacant urban spaces Government {local} Social emotional learning Public Health Youth Aware of Mental Health Program Education Signs of Suicide Teach co in 3 Good Behavior Game and problem- solvmg Sklils Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening Families 10-14 Gatekeeper training Public Health Mental Health First Aid Healthcare Screening combined with care management Healthcare Identify and 51 support people alt-risk Henry Ford Perfect Depression Core {Pre- cursor to Zero Suicide) Best Available Evidence Social Services Crisis Intervention Public Health Active foiio w?up con toct approaches EST for Suicide Prevention Notionoi Suicide Prevention Lifeiine Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide Health ca re improving Mood Promoting Access to v? Social Services Coiioborotive Treatment Justice Coiiohorotive Assessment and Management v? Suic?doii CAMS Intervene to I lessen harms Dioiecticoi Behovioroi Therapy and prevent future risk Attachment?Based Fomiiv Therapy Treatment to prevent re-attempts Health ca re ED Brief intervention with Foiiow-up Visits Social Services v? Postvention Heaithcare 52 Best Available Evidence StandBy Response Service Safe messaging following a suicide Media Guidelines 1/ Public Health Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. Policies, Programs, and Practices to Eupporl} Individuals, Families, 8.: Communities: A Technical Package to Prevent Suicide Prepared lav: Deb Stone, 5CD, MSW, MPH Kristin Holland, MPH Brad Ba rtholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division ofViolence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2016 Comment [Ah From TS: ?support" could be read in a very iimited way - providing services. This seems inconsistent with our emphasis on upstream approaches and work across lthe ecology. Comment lAl: From JM: The term support evokes a specific strategv. Some of what is contained in this package goes beyond support or takes other avenues such as many of the approaches within protective environments. Is there another broader word that oould be used kto characterize this package here? JL Poiicies, Programs, and Practices to Support Individuals, Communities: A Technicai Package to Prevent Suicide is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hourv, MD, MPH, Director Division of Violence Prevention James A. Mercv, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, EL, Crosby, A.E., Davis, 5., and Wilkins, N. Policies, Programs, and Practices to Support Individuals, Families, and Communities: A Technical Package to Prevent Suicide. Atlanta, GA: National Center for Injurv Prevention and Control, Centers for Disease Control and Prevention, 2016. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Strengthen Access to Mental Health Care 16 Create Protective Environments 18 Promote Connectedness 22 Teach Coping and Problem-Solving Skills 24 Identity and Support People At-Risk 28 Intervene to Lessen Harms and Prevent Future Risk 32 Sector Involvement 37 Monitoring and Evaluation 39 Conclusion 40 References 42 Appendix A 50 Acknowledgments [to be inserted later] External Reviewers [to be inserted later] Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (Crosby, Ortega, 8: Melanson, 2011). Self-directed violence may be suicidal or non-suicidall in nature. For the purposes of this document, we refer only to behavior where suicide is intended: 0 Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. Ir Suicide attempt is defined as a non?fora! self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and Worldwide. it contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General 8-: National Action Alliance for Suicide Prevention, 2012,- World Health Organization, 2014). in 291:1(the most recent year of available death data}, suicide was responsible for 42,223 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016dj. in 291d, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1925 in the U5 (Centers for Disease Control and Prevention, 2015dj. Overall suicide rates have increased 24% from 1999 to 2014 (Curtin, Warner, 8: Hedegaard, 2016). [Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?3514} it is the fourth leading cause among persons in their 405 and seventh among persons in their 505 KCenters for Disease Control and Prevention, Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 81 Gfroerer, 2011}. Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts} than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments} or not treated at all (Crosby, Han, et al., 2011]. For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideationj (Ferdon et al., In pressj. Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, is biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological mode -- encompassing multiple levels offocus from the individual, relationship, community, and societal-- is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8: Krug, 2002). Risk and protective factors for suicide exist at each layel. For example, risk factors include: racialjethnic stats Comment IAI: From JM:5eems like too much detail on leading causes, suggest looking for a way to condense. See suggested edit. Comment It might be helpful to balance the description of suicide as {Comment IAI: Add AIIAN stats, there are multiple opportunities for prevention. We don't want readers to think we need to address all of the Lcomplex factors to make a difference. a Comment (Al: From JH: is it worth mentioning terminal disease or ,condition? this to include perpetration too. .. abuse, previous attempt,_vi_olerice_EEti miaatio_rd;_a_nd i 0 Relationship level: High conflict or violent relationships, sense of isolation and lack of social Individuallevei: History of depression and other mental illnesses, hopelessness, alcohol and drug -. support, familyIIOVed one?s history of suicide, financial and work stress Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications) complex with multiple influencers with a positive message about this meaning that {Comment From T5: Could broaden at. 1 Societal level: Availabilitv of lethal means of suicide, unsafe media portravals of suicide, stigma associated with help-seeking and mental illness (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. It is important to recognize that the vast majoritv of individuals who are depressed or who have other risk factors noted, do not die lav suicide. Furthermore, the relevance of each risk factor can varv bv age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General 3: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. Protective factors, or those influences that guard against the risk for suicide, can also be found across the different levels of the social?ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and sLIpportive relationships with partners, friends, and familv; connectedness to school, community and other social institutions; availabilitv of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or guard against a number of risks associated with suicide. Suicide is connected to other forms of violence. Suicide and other forms of violence often share some of the same root causes [Butchart, Phinnev, Check, all Villaveces, 2004; Klevens, Simon, 8: Chen, 2012). For example, in neighborhoods where there is low social cohesion, or where residents don?t support and trust each other, people are at higher risk for suicide l0esai, Dausev, 8f. Rosenheck, 2005) as well as perpetration of child maltreatment [Coulton, Crampton, lrwin, Spilsburv, SI. Korbin, 200?; Freisthler, Merritt, El. LaScala, 2006}, teen dating violence [CapaldL Knohle, Shortt, Kim, 2012), intimate partner violence (Pinchevskv 8t Wright, 2012), and vouth violence (Sampson, Morenoff, SI Gannon-Rowlev, 2002}. Additionally, a lack of economic opportunities and unemplovment are associated with suicide {Luo, Florence, Ouispe-Agnoli, Duvang, 8d. Crosbv, 2011: Reeves et al., 2012), as well as perpetration of child maltreatment Runvan, Wattam, Ikeda, Hassan, 23? Ramiro, 2002}, intimate partner violence {Heise El Garcia-Moreno, 2002; 8: Wright, 2012], sexual violence {Centers for Disease Control and Prevention, 2016c} and vouth violence (Wilson, 2011}. Other shared risk factors for suicide and violence occur at the individual level and include substance abuse, mental health problems, witnessing violence, and a lack of problem-solving skills {Centers for Disease Control and Prevention, 2016a, 2016c, 2016a; U.S. Office of the Surgeon General El National Action Alliance for Suicide Prevention, 2012]. Just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, tonnectedness Increases individuals and communities' resilience to suicide and other forms of violence, including connE'ctedness to one?s communitvk?asile, Hamburger, Swahn, SI Choi, 2013; Borowskv, Hogan, 3: Ireland, 199?; Centers for Disease Comment From TS: lfound this wording somewhat awkward. 2007; Kleirnan, Riskind, Schaefer, 8t Weingarden, 2012; Pinchevsky 8: Wright, 2012; Widome, Sieving, Harpin, 8t Hearst, 2003], school (Basile, Espelage, Rivers, McMahon, 8; Simon, 2009; Capaldi et al., 2012; Carter, McGee, Taylor, Williams, 2007; DeGue et al., 2013; Hong, Kral, Espelage, 81 Allen-Meares, 2012; Losel 8i Farrington, 2012), family [Capaldi et al., 2012; Centers for Disease Control and Prevention, 2016a; if Elgar, Craig, Boyce, Morgan, Vella~2arb, 2009; Maimon, Browning, 8t Brooks?Gunn, 2010; Fiesnick, Ireland, 8t Borowsky, 2004], caring adults (Capaldi et al., 2012; Losel 8i. Farrington, 2012; Maimon et al., 2010], and pro-social peers [Capaldi et al., 2012; Losel EL Farrington, 2012}. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far?reaching consequences for individuals, families, and communities (Dunne, Mcintosh, 8t Dunne- Maxim, 198?; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, unfortunately, survivorship itself is a risk factor for suicide (@rosby Sacks, 200 that the heath con?uence of and death. Suicide attempt survivors those with lived experience] may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity ofthe attempt [Chapman 9 Dixon-Gordon, 200?). Similarly, survivors ofa loved one?s suicide may?xperiencdongoing pain and suffering including complicated grief (Mitchell, Kim, Prigersonu. 8t Mortimer-Stephens, 2004), stigma, depression. anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Julie Cerel, McIntosh, Neimeyer, Maple, St Marshall, 2014; Sudak, Maxim, Carpenter, 2003). The economic toll of suicide is immense as well. [The total lifetime costs associated with nonfatal injuries Comment IAI: From LLD: The Crosby E: Sacks reference is missing from the reference list. I looked it up online and noticed that the 13 million estimate is based on knowing a suicide decedent in the past year. Knowing someone who died by suicide seems potentially broader to me than being directly impacted. and deaths caused by self?directed violence in 2013 were approximately $93.5 billion after adjusting for under-reporting of suicide (Shepard, Gurewich, Lwin, Reed, 8: Silverman, 2016}; The overwhelming! burden of these costs results from lost productivity over the life course, with the average cost per suicide being over $1.3 million [Shepard et al., 2016]. Comment [Ah From TS: Has anyone .1 updated this? lt would be good to check._., 1 Comment [Alt From T5: I don?t think you need to start with "violence" and then fun on suicide. 1 Comment IAI: From LLD: they may also be impacted in other ways ie.g., loss of jamil?ncome, etci 1 Comment From TS: It seems like it would be appropriate to acknowledge the CDC estimate too. Consider using that as a minimum and then say that others have estimated that the costs could be considerably higher after adjusting for Suicide can be prevented. Despite the surrounding suicide, like most public health problems, suicide is preventable Public Health Service, 1999). And while progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research Lunder?reporting of suicide. suggests that suicide will not be prevented by any single intervention taking place in any single setting {Silverman 8i. Maris, 1995; 0.5. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public leg, business, public health, physical and behavioral healthcare, justice, education, labor) (National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014]. Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or i i risk or protective factors for suicide. To be considered for inclusion in the technical package, the i program, practice, or policy selected had to meet at least one of these criteria: a] meta-analyses or i systematic reviews showing impact on suicide; b] evidence from at least one rigorous leg, randomized i controlled trial or quasi-experimental design) evaluation study that found significant preventive i effects on suicide; c} meta?analyses or systematic reviews showing impact on risk or protective factors i for suicide. or d] evidence from at least one rigorous {ego RCT or quasi?experimental design) evaluation i study that found significant impacts on risk or protective factors for suicide. Finally, consideration was i also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of ,l harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in i 51. i a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide ie.g., help-seeking, stigma reduction, depression, connectedness). In terms of the strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the Eievelopmenta?-nature of the evidence base and the use of the best available evidence at a given; time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach ie.g., gatekeeper training} are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The examples provided are not intended to be a comprehensive list of evidence-based programs, policies, or practices for each approach, but rather illustrate models that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. in practice, the effectiveness of the programs, policies and practices identified in this package will be strongly dependent on the quality of their implementation and the communities in which they are implemented. Implementation guidance to assist practitioners, organizations and communities will be developed separately. 1G Comment IAI: From TS: "developmental nature"I doesn't seem correct. Do you mean the ?current status? Context and Cross-Cutting Themes The strategies and approaches that have been included in this technical package represent different levels of the social ecology, with efforts intended to impact the community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide {see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt}. Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security Housing stabilization policies Strengthen access to mental health care 0 Coverage of mental health conditions in health insurance policies Create protective environments I Reducing access to lethal means among persons at-risk of suicide I Organizational policies and culture a Community-based policies to reduce excessive alcohol use Promote connectedness - Peer norm approaches It Community engagement activities Teach coping and problem-solving skills . Social-emotional learning I Parenting skill and family relationship approaches Identify and support people at risk I Gatekeeper training Screening combined with care management Crisis intervention Treatment for people at-risk of suicide Treatment to prevent re-attempts Postvention Safe messaging following a suicide Intervene to lessen harms and prevent future risk The example programs, policies, and practices have been implemented within particular contexts. The social and cultural context of communities is critically important to take into account when selecting strategies and approaches. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their contest. 11 Suicide ideation, attempts, morbidity and mortality vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Barriers to disclosure, help seeking, timely access to quality care, and ongoing support may also vary by population and community characteristics. ideally, the availability of multiple approaches tailored to the economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. These culturally appropriate approaches can then be included in comprehensive strategies to maximize the public health impact on reducing suicide-related morbidity and mortality among individuals and within communities. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. in the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 1 Comment IAI: From TS: was thinking that it might be good to work in a point about how these risks can be directly or indirectly associated with suicide risk because these stressors can exacerbate relationship as well physical and mental health problems. {Comment From TS: i like this clear i and succinct summary. Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old (Luo et al., 2011; Fowler et al., 2015). Economic and ?nancial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and - reduced vs. reduce Comment From LLD: the other packages use "Reductions you may i 1 [Comment From JH:Consistent tense 41 housing expenses, and even the anticipation of such financial stress, can directly orEndirectly] increaseJ.? i i i an individual?s risk for suicide; buffering these risks can therefore, potentially protect against suicide want to consider using similar language for your potential outcomes. {Stack 81 Wasserman, 200?}. For example, strengthening economic support systems can help people pay stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this i . support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing 'l [suiciddL ,5 ill ii Approaches El Economic supports for individuals and families can be strengthened by targeting household financial :3 security and ensuring stability in housing during periods of economic stress. 5" - Strengthening household ?nancial security can potentially bufferthe risk ofsuicide by providing 5 individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other i forms of temporary assistance, livable wages, medical benefits, and retirement and disability .5 insurance to help cover the cost of necessities or to offset costs in the event of disability, are ,5 examples ofways to strengthen household financial security. til I Housing stabilization policies aim to keep people in their homes and provide housing options for El those in need during times of financial insecurity. This may occur through programs that provide 5 affordable housing such as through government subsidies or through other options available to i potential homebuyers such as loan modification programs, move-out planning, or financial 'l counseling services that help minimize the risk or impact of foreclosures and eviction. i 5 Potential LutcomeJ - Reduced suicide rates 0 Lower foreclosure rates Lower eviction rates 13 Reduced emotional distress Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. An examination of variations in US. unemployment bene?t programs across states demonstrated that the impact of unemployment on suicide was offset in those states that provided greater than average unemployment benefits {Cylus, Glymour, Et Avendano, 2014]. Another U.S. study examining the linlt between unemployment and suicide risk using suicide data, length of unemployment, and job losses found that the duration of unemployment, as opposed tojust the loss of job, predicted suicide risk {Classen 3d. Dunn, 2012]. Together, these results suggest that not only should state unemployment benefit programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance] have also shown an impact on suicide. A study by Flavin and Radcliff [2009) examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance {Temporary Assistance to Needy Families and total state spending on suicide rates between 19902000, controlling for a number of suicide risk factors leg, residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn't spending in general that was associated with the reduction but spending on these types of assistance. in terms of lives saved, Flavin 8t Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased their per ca pita spending on these types of assistance $45 per yearKFiavin 8t Radcliff, 200931 Housing stabilization policies. The National Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration of the neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working?aged adults [Houle 8t Light, Another study 14 Comment IAI: From TS: Consider including a point about this being a correlational study and more evaluation Work is needed but it suggests the potential benefits of policies that reach those Iwho are particularly Vulnerable at the times when they are in greatest need. of data from 16 US. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began) to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 1' Comment From .IM: I assume this means that different types of mental Strengthen Access to Mental Health Care I: health problems pose different risks for suicide, but wasn't sure? May want to Lmake the point more explicit JL Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, 3 Comment From T5: Isn't this including substance abuse disorder? it Huang, Stroup, 2015; Owens, 2002}, had risk conferred by mental illnesses ,5 Lapierre, Kim, 31 Turecki, 200d; E. C. Harris 8: Barraclough, 199?; Tyrer, Reed, St Crawford, 2015), previous if Lwould be good to state this explicitly. research indicates that mental Illness is an Important risk factor for suicide (E. C. Harris Barraclotugh, Comment Hi: From TS: I found this 1998} World Health organization, 2014). that Up to 90% 0f people Who die by SUICICIE 1" I sentence confusing Can you reward ta may have had a mental ?llnesd at the time of their deaths {ArsenauIt-Lapierre et al., 2004,- Cavanagh, ,l From LLD: could say: "if a state has a Weaker parity law than the federal parity law leg, includes coverage for some mental health conditions but not others}. then the federal parity law will replace Lthe state law.? Carson, Sharpe, 81 Lawrie, 2003; Isometsa, 2001}. State?level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shem, Bagaiman, 8: Cap, 200?]. Findings from the National Comorbidity Survey indicate that relatively few people in the US. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one ofthe contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health care for people in need is a critical component to suicide prevention {World Health Organization, 2014}. Apart from the treatment benefits, it can also serve to normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns Ii.e., mental health parity}. Benefits and services covered include such things as the number of visits, co?pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. [Federal parity replaces the state law only in cases where the state law prevents the application of the federal parity law ie.g., includes coverage for some mental health 5 conditions but insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Potential Outcomes Increased utilization of mental health services 15 Decreased of mental illnesses 1 Comment IAI: From T5: 1 don't think this is accurate. I was curious about this study and checked out the paper. I think they are saying that this is the cost incurred per suicide prevented. They go on to - Decreased rates of suicide attempts Decreased rates of suicide explain that this is the upper bound and EVldence does not reflect the benefits to non? There is evidence suggesting that coverage of mental health conditions in health insurance policies can SUiEidal i?leiduals- Please con?rm this- If you are going to include this then it will Lhe important to explain it further. i i i reduce risk factors associated with suicide and may directly impact suicide ratesCoverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006) found that 12 months after states enacted mental health parity lows. self-raported use of mental health-care services significantly increased. Moreover, subsequent research by Lang (2013] examined state mental health IaWs and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year andEi cost savings of 513-11 million }per suicide prevented (Lang, 2013). 1? Create Protective Environments Rationale Prevention efforts that focus not only on Individual behavior change help?seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes {Haddom 1980}. Creating environments that address risk and protective factors where individuals live, work, and plav, can help prevent suicide {Dahlberg 81 Krug, 2002; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide and suicide attempts are elevated in certain occupational groups (Han et al., 2016; Mcintosh et al., 2016), therefore, changes to the organizational culture in these occupations, bv war of implementation of supportive policies or evan phvsical modifications to the workplace environment, can change social norms, encourage help?seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not L. Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifving the characteristics of the environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis lBeautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; Crosby, Espitia~ Hardeman, Ortega, Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenwav, 8i. Azrael, 2015; C. W. Runvan et al., 2016}. Approaches The current evidence suggests three bromisin?approaches for creating environments that protect; against suicide. Reducing access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumping from heights provide little opportunitv for rescue and, as such, have high case fatalitv rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that lithe interval between thinking about and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001} and 2) that people tend not to substitute a different method when a highlv lethal method is unavailable or difficult to access lHawton, 200?; Yip et al., 2012]. Therefore, increasing the time interval between the thought and the suicide attempt, for example, bv Comment IAI: From lid: The word promising has certain connotations when using it around evidence. lr?ou sav later that the evidence around this strateg'v is some of the strongest for suicide. These two statements seem somewhat inconsistent. i would use a different word here then promising if vou truly believe there is strong evidence. From JH: suggest deleting the word ?promising? From LLD: since we don't want to infer that these are the onlyr approaches for creating protective environments, I might be inclined to change the word ?promising" to ?potential? or "poSsible" i i I i tor something along those lines. making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening otSuicr'o?e Hotspots. Suicide hotspots, or places where suicides may take place relativelv easilv, include tall structures leg, bridges and cliffs), railwav tracks, and 18 isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help (Cox et al., 2013}. Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating individuals who may be vulnerable andfor impulsive from easy access to lethal means. Such practices may include edUCation and counseling around storing firearms?docked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8t Rivara, 2016; C. W. Runyan et al., E3- 2015). Organizational policies and culture that promote protective environments may be implemented in places of employment. Such policies and cultural values may promote prosocial behavior le.g., asking for help], skill building, changing social norms, referral and access to helping services mental health, substance abUse treatment, financial counseling}, and encourage leadership support from the top down. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) {National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol?involved suicides (Escobedo 8t Ortiz, 2002; Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, takes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one?third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004]. Potential Outcomes Increase in safe storage of means Reduction in suicide attempts Reduction in suicide deaths Increase in help?seeking Reduction in alcohol?related suicide deaths Evidence The evidence for the effectiveness reducing access to lethal means for person at-risk of suicide and other ways to establish protective environments Es some ofthe strongestlin the field Zalsman etal.,20161Comment IAI: From .Il'v1: Consistent with promising used earlier? Is it the strongest? Reducing access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ea. Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removai of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, sadiyeboth the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009). Another form of means reduction involves implementation of safe storage practices. In a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani?Rahbar et al., 2016}. Another program, The Emergency Department Counseiing on Access to Lethai Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre?post quality improvement project, Runyan et al (2016} found that at post-test (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence ofguns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test (C. W. Runyan et al., 2016). Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and heip-seeking (Mishara 81: Martin, 2012). Police suicides were tracked over 12 years and 20' compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 73.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (ratee29.0 per 100,000) (Mishara 81. Martin, 2012) Another example of this approach is the United States Air Force Suicide Prevention Program (AFSPP). AFSPP included 11 policy and education initiatives and was designed to change the culture ofthe Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training}, and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (K. L. Knox, Litts, Talcott, Feig, 8L Caine, 2003). Using a time-series design to examine the impact of the AFSPP program on various violence- related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (K. L. Knox et al., 2003).The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively}, homicide and accidental death (K. L. Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (K. L. Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (K. L. Knox et al., 2010). Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol eseexist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8: Remer, 2009}. 21 1 Promote Connected ness Comment iai: From JM: Seems like a . nuance you can delete from this Il paragraph. You don?t really pick up on ll this later and it probably requires more JL Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, i.e. lack of connectedness, are explanation among the chief causes for suicidality (Durkheim, 1897;1951]. Connectedness is the degree to which an ,3 'Cummnt It?l? Fromm, Awkward individual or group of individuals are socially close, interrelated, or share resources with others (Centers l; .i wording. Consider, while the evidence oclal connections can be formed Within and between Mimi??55L; Comment IAI: From TS: lthink you i' meant to say a "positive? association Lrather than "inverse". . Comment From JM:Thisimp ies i that prosocialbehavioris currently abnormal. Maybe promote is a better word? for Disease Control and Prevention, 2009). multiple levels ofthe social ecology (Dahlberg 0 Krug, 2002}, for instance between individuals leg. peers, neighbors, co-workers], families, schools, neighborhoods, workplace, faith communities, cultural groups, as a community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8t Laud, 2015; Muennig, Cohen, Palmer, 8: Zhu, 2013). Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. ?Nhile the 'l evidence is still being is towards an inversegassociation between social capital measured by social trust, community! neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for From JH: suggest deleting "seek to normalize? and say "Peer norm approaches encourage prosocial Disease Control and Prevention, 2009}. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement can protect against suicide. as help-seeking, reaching out and talking to trusted adults, and peer connectedness. These approaches typically target youth and are delivered in school settings but can also be impIEmented in community settings. 22 JR. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean?up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby Comment IAI: From JH: Cite findings from Let's Connect intervention program? While the intervention didn?t signi?cantly Comment From TS: Is there any reduce suicidal behavior, there 'were signi?cant improvements in connectedness (a protective factor for JR. suicide], evidence on outcomes more proximal to suicide? Did they look for any and not ?nd them? It is important to describe relevant reducing risk of suicide. Potential I - Increase in healthy coping attitudes and behaviors I Increase in referrals for youth in distressed i a Increase help?seeking behaviors ii . . . i: - Posmve perception of adult support li i ii . . i bwdence 5: I I I Current evidence suggests that peer norm approaches and community engagement can reduce risk factors associated with suicidal behaviors. 0 Peer norm Evaluations show that programs such as Sources of Strength cani i Reduction in maladaptive coping attitudes and behaviors Outcomes improve school norms and beliefs about suicide that are created and disseminated by student Lnull effects when they were found. ., Comment IAI: From JM: Not seeing strong evidence here. Are you suggesting that this affects documented risk factors for suicide by reducing stress and increasing physical exercise? I would like to see a stronger evidence statement here. From LLD: you might want to incorporate a phrase in the statement along the lines Lof "which are risk factors for suicide? Comment From T5:They had a i follow Up paper in 2013 that showed i intervention sites felt significantly safer. The effects on crimes was encouraging but not significant. It seems important to mention this too. From LLD: not sure I agree with Tom's last point about mentioning the nonsigniflcant peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high?schools (E metropolitan, 12 rural}, Wyman et al. [2010] found that the program improved peer leaders' i adaptive norms regarding suicide, their connectedness to adults, and school engagement. Peer i i leaders were also more likely than controls to refer a suicidal friend to an adult. Among students, i the intervention increased perceptions of adult support for suicidal youths and the acceptability of seeking help. Perception of adult support increased most in students with a history of suicidal ideation. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained et al., 20105. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,435 lots [or?.3 million square feet} in 4 areas ofthe city. Researchers found significant associated E'eductions in community residents' self?reported stress levels and engagement in more physical exercise han residents in control vacant lot areas. Other benefits included I reductions in firearm et al., 2011} 23 Leffect on crime. Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problemrsolving skills, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014}. Suicide prevention programs thatfocus on life and social skills training are drawn from social cognitive theories (Bandura, 1936i, surmising that individuals lemme moraine and enrimnmerta' and individual leg. hopelessness) characteristics. The literature linking life skills and suicide is robust. The inability to employ adequate seeing-strategies to cope with immediate stressors or identify and find solutions for problems hagee been characterized among suicide attempters {Pollock 8i Williams, 2004). Treatments that include bolstering skills {Goldsmith, Pellmar, Kleinman, 8t Bunney, 2002) and include problem-solving techniques {Ghahramanlou?Holloway, Bhar, Brown, Olsen, Beck, 2012; Townsend et al., 2001} appear to reduce suicidal ideation and attempts more effectively. Prevention programs focused on teaching these skills target youth, parents and families and have been used with both universal and at?risk populations. While many do not target suicidal behaviors directly, these programs strive to train youth and parents life skills to offset the underlying vulnerabilities that contribute to engaging in highvrisk behaviors early in life. Approaches Current evidence provides support for the following two approaches: Social emotional learning programs focus on developing and strengthening communication and problem?solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, Webster-Stratton, 2011]. Parenting skill and family relationship programs are designed to strengthen parenting skills, enhance positive parent?child interactions, and improve children?s behavioral and emotional skills and abilities. Several parenting and family relationship programs have been shoWn in rigorous evaluations to improve resilience and reduce risk factors for various behaviors, including 24 Comment IAI: Speci?c edits needed here. Comment IAI: From LLD: the word ?either" here implies another comparative clause,- should the word "or" be Inserted before the word "modeling?? May want to just delete the word "either? {compare to 505 below) ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse [Cummem From JH: Report 33% (M. S. Knox, Burkhart, 81 Hunter, 2010}. Comment From JH: Suggest ll translating DR and 95% CI for wider if audience. :5 From LLD: agree; this is too technical for a :l wider audience, See Tom's suggestion Lbelow. Jl ll Comment From JH: state if the Potential Outcomes Reduction in suicide attempts and suicide ideation Enhanced knowledge of risk and protective factors associated with suicide I 0 Reduction in suicide risk behaviors li.e., depression, anxiety, conduct problems, substance abuse) . Improve and normalize help-seeking behavior if; control group received any kind of I Enhance social competence and emotional regulation skills :ii itlntewemmn - Enhance problem-solving and conflict management skills ,C?m'mnt Ml: mm m" i ii Comment [Ah From Lancet ll paper has a 2015 publication date. ll The authors provide the absolute and RR i for suicide attempts too and they give a 5 clear way ofthinking about this for 5 an ?of tone pupils, 11 attempted suicide 3: in the control group vs five attempts in 3 reduce suicidal behaviors and associated risk factors. - Social emotional learning programs. The Youth Aware of Mental Health Program is a program developed for teenagers that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide {including knowledge about depression and anxiety} and enhances their problem-solving skills for explanation. dealing with adverse life events, stress, school and other problems. The program includes 3 Kill {Comment Hi: From JH: signi?cant? hours of role-play sessions and interactive workshops combined with a booklet that students i Effect sizeimagnitude of reduction? can keep, educational posters displayed in classroom, and interactive lectures about mental health at the beginning and end of the program [Wasserman et al., 2014). in a cluster- it might be worth using the attempt Evidence 5 There are several programs with evidence that support teaching social, emotional and parenting skills to example so you can provide this schools, Etudentd participating in the TAM program were significantly less likely to have an l' I incident suicide attempt OR 0.45, 0.24?0.85; and severe suicidal ideation i ,l {0.50, 0.27-0.92; ]at the 12-month follow-up compared to the ontrol grou Viw?vv?v' 1-1? Additionallv, related to severe suicide ideation, in the TAM group absolute risk fell byp.50%- land relative risk fell by 519.6% EWasserman et al., 2014} Signs ofSuicide {505} is another school-based prevention program for students aged 13-17. The program includes guided classroom discussions about suicide and depression. As part of the program, students are screened for depression and suicide risk and referred for professional help as indicated. The program is designed to increase knowledge about suicide and risk factors associated with suicidal behavior as Well as improve and normalize help- seeking behavior (Schilling, Aseltine, St James, 2016). in a randomized controlled trial, SOS was shown to suicide attempts at 3-months post intervention amongli participating students compared to [:ontrol students. The SOS program also increase students' knowledge of how to get help for themselves or friends for depression and/or suicidal thoughts, and favorable attitudes toward help?seeking. 505 participants with a lifetime history of suicide attempt were also less likely to report planning a suicide in the 3 months following the program compared to ?ower-risk participants]5chi ling et al., 2016}. Finally, the Good Behavior Game {636) is a classroom?based program for elementary school i .4 .I [Comment From JH: signi?cant? 0 r? ,4 children aged 6-10; it represents an example of upstream suicide prevention programming The program uses a team?based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the (336 is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior {Wilcox et al., 2003). in an outcome evaluation of the 636, first graders assigned to 636 reported half the adjusted odds of suicidal ideation and suicide [attemptd The beneficial Jr effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 686 effect on attempts was less robust in some adjusted models including caregiver mental health. in the hecond cohort 10f 686 students, neither ,r'l 3" suicidal ideation nor suicide attempts were significantly different between 636 and the i l' I control interventions (Wilcox et al., 2008} ifhis finding likely arose clue to the lack of implementation fidelity] and pointed to the need for (33-5 to be delivered with precision, consistency, and teacher support. (336 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide {Kellam et al., 2008}. Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors, in youth by improving protective factors such as responsive and positive parent?teacher?child interactions and relationships, emotion self?regulation and social competence {all protective factors for suicide] (Herman et al., 2011]. The program includes 9? 20 sessions offered in community?based settings religious, recreation centers, mental health treatment centers, and hospitals}. Several studies have demonstrated the effect of the li? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, 8t Beauchaine, 2011; Jamila Reid, 8: Stoolmiller, Elli-03). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year followrup [Reid, Webster?Stratton, 8; Hammond, 2003; C. 8; HammondComment IAI: From define control condition Comment [Ah From JH: Lower-risk? Do you mean students with no prior reported Ihistory of suicide attempts? Comment From LLD: Jeff had questions about what is meant by upstream. You are referring to primary prevention but some of the other examples YAM are also primary prevention. Suggest just deleting this Lstatement. tommentji?: From JH: compared Comment IAI: From .IH: Suggest stating clearly there are two separate cohorts of students receiving the intervention. How Ldo these cohorts differ (time, location}? Comment From TS: it is important to help the reader to understand that this study looked at the suicidal ideation when the ?rst graders were age 19-21. This is m_otclear currently. Comment From is this the authors? comment or your interpretation Lof the data? Please clarify. I 1997; C. Webster?Stratton, Reid, Sn Hammond, 2001). The program demonstrated greater . gamma?. lAl: benefits as the dosage of the intervention increased (Herman et al., 2011). 3 N50 de?ne externalizing behaviors. i Additionally, Strengthening Families Iii-l4 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting vouths? interpersonal and problem?solving skills; and creating familv activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guvll, 8i. Day, 2002]. Strengthening Families has been shown to ecrease externalizing hehaviord1 alcohol use, and drug use if among vouth participants gangreductions in depression, participating families [Spoth 2? Identify and Support People AtaRisk Rationale In order to decrease suicide, Littention to people at increased or high risk is necessary as these individuals tend to experience suicidal behavior at higher than average rates.]These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously; individuals who are institutionalized, have been victims of violence, or are homeless; and members of certain racial and ethnic minority groups. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at?risk or vulnerable groups, customizing services to make them accessible and maieta?ia?ingengaged in care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. People who are disadvantaged face social and economic issues that may adversely affect their ability to respond to the treatments or advice that are offered. Approaches The following three approaches focus on identifying and supporting people at increased risk. 0 Gatekeeper training is designed to train teachers, coaches, providers and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training is typically implemented in schools to identify at?risk youth and within health care settings to identify adults {and youth}. 0 Screening combined with care management and overall continuity of care has been used in primary care and behavioral health care settings to assure that people who may be at high-risk of suicide are identified and receive ongoing treatment as needed, particularly after inpatient discharge and other transitions within the healthcare system so they don?t ?slip through the cracks?. These approaches typically employ screening for depression and/or suicide combined with collaborative treatment planning between patients and their providers and patient follow- up. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, or text messaging. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, 28 Comment Hi: From TS: This seems awkward as written. it might be better to say ",..attention to people with speci?c crisis interventions can put space or time between an individual who may be considering suicide :Cnmmem From suggest i shortening this introductory statement i about the evidence and making it i consistent with the other sections. If you need to provide caveats, then you might want to take a look at the TP for Lexample wording. I Comment From TS: Add cites here. Also this wording is awkward because it could be read as suggesting that those who chose to use the hotline are different and harmful behavior. JL Potential Outcomes Reduction in suicide attempts {Comment From JH: compared to? I - RedUction in suicide deaths - Increased identification of individuals at-risk for suicidal behavior I Increased at?risk individuals in treatment i from others at risk who did not. Maybe 0 Increased community members trained to identify at-risk Individuals You could add the you? up period to the II Increased referrals for health care i I end -they have decrease suicidal ll Lthoughts and behavior {Comment From LLD: wording is a bit ii i awkward here II I Comment From .IH: signi?cant? Is it possible to report magnitude of intervention effects? Comment From JH: Overall mental health or specific facets depression. Evidence [There is evidence that community gatekeeper programs are successful in reducing suicides and suicide attempts but the efforts must be maintained [Substance Abuse and Mental Heaith Services Jr,- Administration, 2014]. However, there is limited evidence for effectiveness giscreerling programs, but ll t' at the same time, standard principles for public health screening make them promising [Pena 8t Caine, 2006]. The number of studies evaluating crisis intervention services is iimited, but a few studies do 35?; indicate that those who use the hotline services have decreased suicidal thoughts andiehavio Fifi; Bit-l? Please clarify if; {Comment From .IH: Native ll American? I Gatekeeper training. One example of gatekeeper training is the Mental Health First Aid program. This program is designed for the lay public and consists of three weekly sessions of three hours each. Participants learn the of people in mental health crises andfor in the early stages of mental health problems those experiencing suicidal thoughts and behavior, acute stress reaction, panic attacks and acute behavior, and depression, anxiety, and disorders], possible risk factors, and where and how to getEevidence-based effectivihelp [Kitchener Iorm, 2004]. In a randomized controlled trial of 300 participants of; MHFA, others, greater likelihood of advising people to seek professional help, improved concordance with health professionals about treatments, and decreased stigmatizing attitudes. Additionally, the intervention resulted in improved Enental health]of the participants themselvesrAI-l?result?s (Kitchener Jorm, 2004]. Additional research rigorously evaluating MHFA for its impact on the?fi?est?ai?dintervention recipientsi?t?hemseivesand suicidal behavior is needed (Kitchener 3t Iorm, 2006]. Gatekeeper training has also been a core part of all Garret tee Smith Suicide Prevention Program which is in place in 49 states and 43 Eribeg. A multi-site evaluation assessed the: 29 connection between community gatekeeper training and a reduction of suicide attempts and deaths by comparing the change in suicide mortality rates and nonfatal suicidal behavior among . the population aged 10?24 in counties implementing GLS trainings, with the trajectory observed fl in similar counties that did not implement these trainings. Counties implementing GLS trainings i had signi?cantly lower youth suicide rates the year following the training implementationivae?i, i 3:793} {Walrath, Garraza, Reid, Goldston, 8.: McKeon, 2015), This finding represents a decrease ll of 1 suicide death per 100,000 among yogh_s10 to 24 yeargLage aids or the avoidance of 3 approximately 23? deaths in this age group between 200? and 2010. Counties implementing GL9 i program activities also had significantly lower suicide attempt rates among youths 16 to 23 years i of age in the year following implementation of the 615 program than did similar counties that i did not implement GL5 program activities [4.9 fewer attempts per 1000 Garraza, 1 llv'alrath, Goldston, Reid, McKeon, 2015]]. More than 79,000 suicide attempts may have been averted during the period studied following implementation of the GL3 program. 0 Screening combined with care management and overall continuity of care. [The Henry Ford Perfect Depression Care program was the precursor to Zero Suicide, and its overall goal was to eliminate suicide. More broadly, though, the aim was to completely redesign depression care delivery to achieve breakthrough improvement in quality and safety by focusing on six aims: effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program developed concrete measures to assess progress on each of these aims and began with screening and 355essment of each patient for suicide risk with coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the Henry Ford Perfect Depression Core {Pris-cursor to Zero Suicide) program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years prior to the intervention [1999 and 2000] to the intervention years (2002-2009). During this time period, the suicide rate fell E. Coffey, 2006; C. E. Coffey, Coffey, 8A Ahmedanl, 2013). Further, suicid rates also declined antennae; and received mental health specialty services. However, for those HMCJ members who accessed 5 only general medical services as opposed to specialty mental health services, the suicide rate increasedliivl. Coffey, Coffey, Ahmedani, 2015}. I Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the National Suicide Prevention trfeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 330 of those completed a follow-up assessment between 1 and 52 days after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also 30 Comment IAI: From JM: I think you need to provide a little more context on Henry Ford so the readers understands in what kind of population the reductions were experienced. I assume from the end that Henry Ford is some sort of HMO, but just a little more information would help clarify this description to readers. Comment IAI: From JH: how much? found that participants experienced significant Elecreases} in suicidality over the course of the FromJH'. Report telephone session, and that levels of hopelessness and pain continued to decrease magnitude 0i [after their initial Kalafat, Harrismunfakh, 8: Kleinman, 2007]. xx'it?Jummem From JH: how long? in another study, this time employing a randomized controlled trial, Gould, Cross, Pisanl, Munfakh, and Kleinman {2013) assessed the impact of the Applied Suicide intertwention Skills Training (ASIST), a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. The training was evaluated across the NSPL network of hotlines GVEF the period 2008*2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors trained in were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call to the hotline. Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 20131. 31 Comment IAI: From LLD: simplify statement and make it consistent with the other introductory statements about approaches. ll'ou may also want to take a look at the introductory statements to kthis section in the other TF5. Intervene to Lessen Harms and Prevent Future Risk Jk Comment Ml: From LLD: Suggest ?ashing this description out a bit. You may want to mention how these are typically delivered one on one or group formats; typical number of Lsessions, location, etc}. Rationale Individuals who have experienced mental health challenges, suicidal ideation, andfaewho have made suicide attempts aedqior haveengaged in non-suicidal self-injury are at increased risk of suicide (US. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012). Risk of suicide can also increase among those who have lost a friend, family member, co?worker, or other acquaintance i to suicide lPitman, Osborn, King, St Erlangsen, 2914]. Exposure to sensationalized or uninformed i i reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion [Ettersdorfer Er Sonneck, 1998; Niederkrotenthaier 3: Sonneck, 2007). Approaches EA broad array of approaches to lesson harms and redUce future risk of suicide among those at increased risk include the provision of mental health care and improved continuity of care, improving linkage to care through active post-discharge planning and follow-up that decreases barriers to ongoing therapeutic support, increasing connectedness to supportive others, addressing bereavement, and framing cemmunications to emphasize resilience, decrease negative affect, and to prevent contagion] I [l'reatment for people at-rlsk of suicide ]typically includes various forms of delivered by licensed providers to help individuals with mental health problems and other risk factors for suicide with problem-solving, impulsivity and emotion regulation. a Treatment to prevent re?attempts. These approaches typically include follow?up contact and use diverse modalities leg, home visits, mail, telephone, e-mail] to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one? on-one interpersonal therapy andfor group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296-2596 reattempt within a year, and 396-994.. of attempt survivors die by suicide within 1 to 5 years of their initial attempt {lnagaki et al., 2015} I Postvention approaches are implemented offer a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief {Szumilas St Kutcher, 2011). 32 Potential buxom-as} {Comment IAI: From .IHL add something about contagion? Safe messaging following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide i f: contagion. Therefore, responsible and safe reporting may help prevent suicide and suicide contagion. Reduction in mental health?related sequelae Increase connectedness Improved coping skills Improved messaging following suicide Reduction in re?attempts Evidenco The evidence addressing strategies to lesson harm and prevent future risk of suicide includes the evaluation of effects of specific approaches on risk and protective factors as well as suicide-related mortality. However, because the evaluation of suicide-related mortality requires large sample sizes and extended follow?up, much of the evidence in this area primarily focuses on risk and protective factors. Treatment for people at?risk of suicide. There are a number of treatments with evidence of impact on risk and protective factors for suicide. One example is the Improving Mood- Promoting Access to Coiiaborative Treatment program. aims to prevent suicide among older primary care patients by reducing suicide ideation and depression in primary care settings. It facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase] by a depression care manager {Hunkeler et al., The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideatian over 24~months of follow-up {Hunkeler et al., Unutzer et al., 2006?; relative to patients who received care as usual. Another example is Collaborative Assessment and Management which is a therapeutic approach for suicide-specific assessment and treatment of patient?s suicide risk. This flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. CAMS sessions are collaborative and involve constant 33 IAI: From T5: thinks. word is batientlabout what is and is not working with the ultimate goal of enhancing the therapeutic I, missing? "patient input"? alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported' In 6 correlational studies (lobes, 2012), In a variety of inpatient and outpatient settings Comment Unclear- and in one RCT with several additional RCTs under wav. CAMS has been associated Accurdmg tothe armle?.thl5 program was evaluated In five Countries. Please state sIgnificant improvements in suicidal ideation, overall distress, and feelings of hopelessness at 12 month follow-up among a community?based sample of outpatIents. (Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DIST) and Attochmenb?asea Family Therapy a multicomponent therapv for individuals at high risk for suicide and who may struggle with impulsivit'iir and emotional regulation. The components of BET include IndIvIdual therapv, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two vear follow up than women receiving communitv treatment [23% vs reqUIred less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006). a program for adolescents aged 1248 and is designed to treat clinicallv diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxietv {Diamond at al, 2010] A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantIi,r greater improvement in suicidal ideation over 24 weeks of follow up than dId adolescents assigned to enhanced usual care. Additionallv, a significantly higher percentage ofABFi" participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (59.2% vs. and at 24 weeks [82.1% vs. 46.2mm) [Diamond et al., 2010). I Treatment to prevent re-attempts. Several strategies that aim to prevent re?attempts have demonstrated impact on reducing suicidal behavior. For example, Emergency Department Brief intervention with Follow?up ILi?isi?ts is a program that involves a one?hour discharge Information session that addresses suicidal behavior, distress, risk and protective factors, alternatives to suicidal behavior and referral options combined with nine follow?up contacts over 13 months (at 1,2,4, 11 weeks and 4, 6,12,18 months]. Follow- up contacts are either conducted bv, - phone or through home visits according to a specific time line for up to 18? months. A randomized :l controlled trial that enrolled suicide attempters from eight hospital emergencv departments Its?months was associated with significantlv fewer deaths from suicide relative to a treatment- as usual group versus respectivelv?W] (Fleischmann et al., 2008). 34 r{IT-rattlmenI! IAI: From .IH: De?ne [follow- ups greater than year] [Comment IAI: From JH: De?ne Another example of treatment to prevent re-attempts involves active foilow-up contact i [Comment Ml: From T5:?found? is approaches such as postcards, letters, and telephone calls?are intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically leg, or every few months in the first 12 ll: months post-discharge with some programs continuing contact for two or more yea rs]. i analysis conducted by lnagaki et al. [2015}, interventions to prevent repeat suicidal behavior in i Merits admitte_dtgan emergency department for suicide attempt wereiheseaapreac?hes-have been found redUCe reattempts by ideation and attempts lHassanian-Moghaddam, Sarjami, Kolahi, 84 Carter, 2011; Wang et al., 2016). Finally, Cognitive Behavior TheropyforSuicide Prevention is an example ofa therapeutic approach to pretrent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related dif?culties) leading up to and following the suicidal event; safety plan development: skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A i randomized controlled trial feared?of designed to prevent rapeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual [Brown et al., 2005]. Pastvention programs are implemented with the goal of providing support to survivors of others? suicide to reduce their own risk of suicide. One example of a postvention program, StandBv Response Service [Stond5y}, provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs 35 .- 1 Comment IAI: From TS: Don?t you mean ?Recommendations for Reporting on Suicide"? Please cite: arts .or The media did not want ?guidelines" and they reacted negativelyI to the idea of Lexternal guidelines. Comment From TS: Are Ivou sure Campbell, 2008}. that the 31 reduction Wasjust on the subway system? i thought that was Safe messaging following a suicide. Safe messaging after a suicide can help assure that reporting of I the event is done in such a way to reduce risk to consumers of news media and other messaging gnationallv. Please confirm. who mav be particularly vulnerable. One way to ensure safe messaging following a smcide encourage that reporters adhere to media gurdehnesfor reporting on Euicides} Reports that are commlm -me Other . . . . . . .I evaluations available? both mclusrve of sumde preventton messages, stones of hope and resdlence, risk and protective i Comment IAI: From TS:Thomas also (Visser, Comans, EL Scuffham, 2014}. In a study by Visser et al. [2014), StondBv clients were significantly less likely.r to be at high risk for suicidalitv than a suicide bereaved comparison group who had not had contact with the Stond?y program (48% and 64% respectively]. Additionallv, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors self?refer for services] (J. Cerel EL 1'5 if ofgoocl reportingreport that suggested benefits b' .rc ch.or content b' rc JL factors, and links to helping resources hotline} and that avoid sensationalizing events or reducing suicide to one cause can help reduce the likelihood of suicide contagion. The most compelling evidence supporting the effect of media guidelines on reduction in suicides comes from Austria. After a sharp increase in suicides on the Viennese subwav, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting?i resulted in a significant reduction of 81 the Viennese 3.: Sonneck, 200?] I f: I 35 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, health care insurers and providers, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Access to Mental Health Care. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 37 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People At?Risk and to Lessen Harms and Prevent Future Risk. The intensity and activities of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health care, social services, and justice sectors can work collaboratively to support individuals at high- risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 38 Comment From This description seems incomplete comparted to the Monitoring and Evaluation i description of YRBS Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are necessary for program implementation as planning, implementation, and assessment all rely on accurate measurement of the problem. i Surveillance data helps researchers and practitioners track changes in the burden ofsuicide. Surveillance systems exist at the federal, state, and local levels. it is important to assess the availability of surveillance i data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National 'v'iolent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. for example, is a state-based surveillance system that combines data from death certi?cates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, 3: Crosby, 2016}. The National Electronic Injury Surveillance System-All Injury Program provides 5 nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc), age, racefethnicity, sex, disposition (where the injured person goes i i i i I when released from the Emergency Department). In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health?risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013}. The mass data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Health is an annual nationwide survey of individuals aged 12 years and older that provides national and state?level estimates of drug use and mental health-related issues, including suicide ideation and suicide Ettempt? It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long? term outcomes are an important part of program evaluation. 39 I additional research is needed to understand the impact of prevention Earogramtion preventing Euicide, ]as opposed to merely examining the effectiveness of programs to impact risk factors associated with suicide. More research is also needed to examine the effectiveness of upstream and community-level The evidencevbase for suicide prevention has advanced greatly over the last few decades. However, .- .r strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the Esffectiveness of combinations of the strategies Jand approaches included in this package. Most "x existing evaluations focus on approaches implemented in there is potential to understand the synergistic effects within a comprehensive prevention approach. dditional research is needed to understand the extent to which combinations of strategies and approaches result in greater I reductions in suicide than individual programs, practices, or policiES] r; Lprior sentence. Comment IAI: From TS: lfound this somewhat awkward. The idea of suicide struggling seemed off and the point is not entirely clear. I think you can be more specific and reference an earlier point in ?the media recommendations section. 4? {Comment IAI: From JH: is jargon. Comment IAI: From TS: Should this be programs, policies, and practices to be Lconsistent? Comment IAI: From JH: including "(Comment [41: From JH: Jgreatjciointl Comment From JH: redundant with ?1 Comment From TS: lthink you can rop this and just use the more escriptive text. You don?t need both. You ould describe this as a range. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. And-w?hile suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide, on a daily basis-?at home, at work, and in our communities. Suicide and suicide attempts are therefore public health issues of societal concern. Fortunately, like many public health problems, suicide is preventable, and fortunately more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release ofthe first world report on suicide, and more timely surveillance data, to name just a few examples. Unfortunately and unlike most other public health problems, suicide still struggles against stigma, shame, and secrecy related to help?seeking, mental illness, being a survivor, or someone with lived experience; misplaced fear of asking someone about their risk of suicide (versus the fear and consequence of not asking}, and fear of taking up certain strategies known to be effective but perhaps unpopular; misinformation about suicide preventability, and disproportionate funding given the public health burder{ Suicide also struggles against the right degree of awareness where too much information, for example by well-meaning reporters and others, may actually do harm. includes strategies and approaches that ideally would be used in a comprehensive fashion, in combination??in a multi?ievel, multi?sectoral way. This technical package includes strategies and]; approaches targeting preventior? leg, social emotional learning for children and youth}, asll well as strategies focused more cognitive behavioral treatment to prevent re- It includes fmiversal, selective, and indicated strategies, or]strategies that focus on the whole 1 population regardless of risk to strategies that focus on those groups at highest risk. lmportantly, this technical package extends the bounds ofthe typical prevention strategies to consider approaches at the 4G Esoter levels of the social ecologsl policies to stabilize housing and communitv engagement initiatives. . . a? . While the evidence base continues to bebuiltemerge, the collection of programs, policies, and practices laid out here are available for implementation now. And in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. In closing, and in keeping with a message of resilience as spoken bv those with lived experience, 'hope, help, and healing is possible.? 41 Comment IAI: From JM: Yes. this is jargon, so the helps, but you might want to ?gure out a wav to say this without referring to the outer levels of _the social ecology. Comment This sets vou up for people to sav oh yeah what AIIAN, transgendered people, prisoners Letc. References Arsenault-Lapierre, G., Kim, C., Turecki, G. (2004). diagnoses in 3275 suicides: a meta-analysis. BMC 4, 37. Bandura, A. (1986}. Sociaifoundations of thought and action: A sociai cognitive theory: Prentice-Hall, Inc. Basile, K. C., Espelage, D. L., Rivers, McMahon, P. M., E: Simon, T. R. (2009). The theoretical and empirical links between bullying behavior and male sexual violence perpetration. Aggression and vioient behavior, 14(5), Basile, K. C., Hamburger, M. E., Swahn, M. H., at Choi, C. (2013). Sexual violence perpetration by adolescents in dating versus same-sex peer relationships: differences in associated risk and protective factors. Western Journai of Emergency Medicine, 14W), 329. Beautrais, A. L. (2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. doi:10.1080fDOUA867010060501 Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Beyer, K. M., Layde, P. M., Hamberger, L. K., 8: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women, 21(1), 49-64. doi:10.1177f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppl 1, i?-ill. Borowsky, l. W., Hogan, M., Ireland, M. (1997). Adolescent sexual aggression: risk and protective factors. Pediatrics, 100(6), E7. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. D., Ten Have, T. R. (2011]. A difference- in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11], 1296-1306. Brown, G. K., Ten Have, T., Henriques, G. R., the, S. X., Hollander, J. E., Beck, A. T. (2005]. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?570. doi:10.1001ijama.294.5.563 Butchart, A, Phinney, A., Check, P., Villaveces, A. (2004). Preventing Vioience: A Guide to impiementing the Recommendations of the World Report on Vioience and Heaith. Geneva, Switzerland: World Health Organization. Capaldi, D. M, Knoble, N. H., Shortt, J. W., 8: Kim, H. K. (2012]. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., 8-: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adoiesc, 30(1), 51?62. Cavanagh, J. T., Carson, A. J., Sharpe, M., 31. Lawrie, S. M. (2003}. autopsy studies of suicide: a systematic review. Med, 33(3), 395?405. Centers for Disease Control and Prevention. (2009}. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versionnaJde Centers for Disease Control and Prevention. (2016a). Child Maltreatment: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016b). Elder Maltreatment: Risk and Protective Factors. Retrieved from Available at: 42 Centers for Disease Control and Prevention. (2016c). Sexual Violence: Risk and Protective Factors. Retrieved from Available at: Centers for Disease Control and Prevention. (2016d). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: Centers for Disease Control and Prevention. (2016a). Youth Violence: Risk and Protective Factors. Retrieved from Available at: Cerel,1., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel,1., McIntosh, 1. L., Neimeyer, R. A., Maple, M., 8; Marshall, D. (2014). The Continuum of ?Survivorship": Definitional Issues in the Aftermath of Suicide. Suicide and lace-threatening behavior, 44(6), 591-600. Chapman, A. L., S: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Cherpitel, C. 1., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiism: Ciinicai and Experimentai Research, 28(5 SUPPL.), 185-235. Classen, T. 1., 8: Dunn, R. A. [2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. doi:10.1002lhec.1719 Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 57(10), 1524?1526. Coffey, C. E., Coffey, M. 1., 3a Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M, Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., lobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28{11), 963072. doi:10.1002fda.20895 Coulton, C. 1., Crampton, D. 5., Irwin, M., Spilsbury, 1. C., 8: Korbin, 1. E. (2007). How neighborhoods influence child maltreatment: a review of the literature and alternative pathways. Chiid Abuse Negi, 11111142. Cox, G. R., Owens, C., Robinson, 1., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, 1. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 81'. Loza no, B. (2013). Alcohol and suicide. Aicohoi; Science, Poiicy and Pubiic Health, 190. Crosby, A. E., Han, 0., Ortega, L., Parks, 5. E., 84 Gfroerer,1. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 1:13 Years United States, 2003*2009. MMWR, 60(No. 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directeo' Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Cunningham, P. 1. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff {Miilwood}, 28(3), w490-501. Curtin, S. C., Warner, M., 8L Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus,1., Glymour, M. M., Avendano, M. {2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am 1 Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 43 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano (Eds), World Report on lv?iolence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. DeGue, S., Massetti, G. M., Holt, M. K., Tharp, A. T., Valle, L. A., Matjasko, J. L., 8: Lippy, C. (2013). Identifying links between sexual violence and youth violence perpetration: New opportunities for sexual violence prevention. of violence, 140. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Desai, R. A., Dausey, D. J., 84 Rosenheck, R. A. {2005). Mental health service delivery and suicide risk: the role of individual patient and facility factors. Arnl 152(2), 311-318. doi:10.1176lappi.ajp.162.2.311 Diamond, (3. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. [2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1987'). Suicide aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8L G. Simpson, trans). Glencoe, it: Free work published 1897). Elgar, F. J., Craig, W., Boyce, W., Morgan, A., 3i. \lella?Zarb, R. (2009). Income inequality and school bullying: multilevel study of adolescents in 37 countries. 1 Adolesc Health, 45(4), 351?359. Escobedo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accia' Anal Prev, 34(5), 6894594. Etzersdorfer, E., 81 Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, 6. A., 8: lskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Fowler, K. A., Gladden, R. M., \lagi, K. J., Barnes, J., El. Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 311?316. Freisthler, B., Merritt, D. H., LaScala, E. A. (2006). Understanding the ecology of child maltreatment: a review of the literature and directions for future research. Child Maltreat, 11(3), 263-280. doi:10.1172l1077559506289524 Ghahramanlou-Holloway, M., Bhar, S., Brown, G., Olsen, (3., 8: Beck, A. T. (2012). Changes in problem-solving appraisal after cognitive therapy for the prevention of suicide. Medicine, 42(06}, 1185- 1193. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Godoy Garraza, L., Walrath, C., Goldston, D. B., Reid, H., 81. McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. 44 Goldsmith, S. K., Pellmar, T. C., Kleinman, A. M, 82 Bunney, W. E. (2002). Reducing suicide: A nationai imperative: National Academies Press. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 333-352. doi:10.1521fsuli2007373338 Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 707-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Public Heaith Reports, 95(5), 411-421. Han, B., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 81 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 176-186. Harris, E. C., St Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205-228. Harris, E. C., Barraclough, B. (1998). Excess mortality of mental disorder. Br,f 173, 11-53. Harris, K. M, Carpenter, C., 8: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 3L Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self?poisoning. British Joarnai of 198(4), 309?316. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. doi101021702215511028.514 Heise, L., 81 Garcia?Moreno, C. (2002). Violence by Intimate Partners. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), Worid Report on Vioience and Heaith (pp. 87'121). Geneva, Switzerland: World Health Organization. Herman, K. C., Borden, L. A., Reinke, W. M., 8: WebsteraStratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co?Occurring Internalizing Q, 26(3), 189-201. Hong, J. S., Kral, M. J., Espelage, D. L., Allen?Mea res, P. (2012). The social ecology of adolescentainitiated parent abuse: a review of the literature. Chiid Hum Dev, 43(3), 431-454. 011~0273dy Houle, J. N., 81 Light, M. T. (2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073?1079. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ, 332(7536), 259-263. Inagaki, M., Kawashima, T., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. JAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Isometsa, E. T. (2001). autopsy studies--a review. Eur 16(7), 379-385. Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. doi:10.11117j.1943? 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8i. Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Aicohoi Ciin Exp Res, 33(12), 2124-2133. 45 Kaplan, M. 5., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, 5. 6., Brown, C. H., Poduska, J. M., Ialongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. C., Demler, 0., Frank, R. G., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515-2523. Kitchener, B. A., 8: Jorm, A. F. (2004). Mental health first aid training in a workplace setting: a randomized controlled trial BMC 4, 23. Kitchener, B. A., 8: Jorm, A. F. (2006). Mental health first aid training: review of evaluation studies. Aust 21 40(1), 6-8. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi:10.1027l0227? 5910/a000136 Klevens, J., Simon, T. H., at Chen, J. (2012]. Are the perpetrators of violence one and the same? Exploring the co- occurrence of perpetration of physical aggression in the United States. lnterpers Violence, 27(10), 1937?2002. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 1375. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Knox, M. 5., Burkhart, K., 81 Hunter, K. E. (2010). ACT against violence pa rents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 23u83. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow?up ofdialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 752-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Luo, F., Florence, C. S., QuispeuAgnoli, M., Ouyang, L., Crosby, A. E. (2011). Impact of business cycles on US suicide rates, 1928-2007. Am Public Health, 101(6), 1139-1146. doi:10.2105fajph.2010.300010 Maimon, D., Browning, C. R., 81 Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. doi:10.117710022146510377878 Mark, T. L., Shern, D. L., Bagalman, J. E., at Cao, Z. (2007}. Banking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, 0. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group 17 States, 2012. Morb Mortal Wkly Rep, 65(25), 641-645. doi:10.15585lm mwr.mm6525a1 Miller, M., Warren, M., Hemenway, D., Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. 46 Mishara, B. L., 8: Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8L Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., 81. Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc SciMed, 85, 18-26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Biaeprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. {2015). Responding to Grief, Trauma, and Distress After a Suicide: US. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., 8: Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust 21 41. Olfson, M., Gerhard, T., Huang, (2., 5., 81 Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172?1181. Owens, D. (2002). Fatal and non?fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Pena, J. 3., 8L Caine, E. D. (2006). Screening as an approach for adolescent suicide prevention. Suicide Life Threat Behav, 35(6), 614-637. doi:10.1521isuli2006366614 Perron, 5., Burrows, 5., Fournier, M, Perron, P. A., Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health, 103(7), 1235-1239. Pinchevsky, G. M., 8: Wright, E. M. {2012). The impact of neighborhoods an intimate partner violence and victimization. Trauma Violence Abuse, 13(2), 112-132. Pirkis, J., Too, L. S., Spittal, M. J., Krvsinska, K., Robinson, J., 8i. Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994? 1001. Pitman, A., Osborn, D., King, M., 31 Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 36-94. Pollock, L. R., Williams, J. M. (2004). Problem?solving in suicide attempters. Med, 34(1), 163-167. Reeves, A., Stuckler, D., McKee, M., Gunnell, 0., Chang, S. S., Basu, S. (2012). Increase in state suicide rates in the USA during economic recession. Lancet, 380(9856), 1813-1814. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Resnick, M. 0., Ireland, M., Borowskv, I. (2004). Youth violence perpetration: what protects? What predicts? Findings from the National Longitudinal Study of Adolescent Health. JAdalesc Health, 35(5), 424 e421? 410. Rowhani-Rahba r, A., Simonetti, J. A., 8-1 Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. doi:10.1093fepirevimxv006 47 Runyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., 31. Novins, D. (2016}. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West} Emerg Med, 1 8-14. Runyan, 0., Wattam, C., Ikeda, R., Hassan, F., 8; Ramiro, L. (2002). Child abuse and neglect by parents and other caregivers. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 81 R. Lozano World Report on Violence and Health (pp. 59-86}. Geneva, Switzerland: World Health Organization. Sampson, R. J., Morenoff, J. 0., 3t Gan non-Rowley, T. (2002). Assessing "Neighborhood Effects": Social Processes and New Directions in Research. Annual Review of Sociology, 28, 443-478. Schilling, E. A., Aseltine, R. H., Jr., 8: James, A. (2016). The SOS Suicide Prevention Program: Further Evidence of Efficacy and Effectiveness. Prev Sci, 17(2), 157-166. Shepard, D. 5., Gurewich, 0., Lwin, A. K., Reed, G. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., St Maris, R. W. (1995}. The prevention of suicidal behaviors: An overview. Suicide and life- threatening behavior, 25(1), 10-21. Simon, O. R., Swann, A. C., Powell, K. E., Potter, L. B., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide the Threat Behav, 32(1 Suppl), 49-59. Spoth, R. L., Guyll, M., 81. Day, S. X. (2002}. Universal family-focused interventions in alcohol-use disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 223. Stack, 5., 8L Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 32(1), 103?112. Substance Abuse and Mental Health Services Administration. (2014). Report to Congress: Garrett Lee Smith Suicide Prevention Program. Rockville, MD. Sudak, H., Maxim, K., St Carpenter, M. {2008}. Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136?142. Szumilas, M, at Kutcher, 5. (2011). Post~suicide intervention programs: a systematic review. Can Public Health, 102(1), 18?29. Townsend, E., Hawton, K., Altman, D. G., Arensman, E., Gunnell, 0., Hazell, P., . . .Van Heeringen, K. (2001}. The efficacy of problem?solving treatments after deliberate self-harm: meta?analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Medicine, 31(6), 979?988. Tyrer, P., Reed, G. M., 8.: Crawford, M. J. (2015}. Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717?726. US. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. US. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006}. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10}, 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014}. Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journalof Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015}. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. doi:10.2105lAJ PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. 48 Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(9977), 1536-1544. Webster-Stratton, C., Hammond, M. [1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 65(1), 93-109. Webster-Stratton, C., Reid, M. 8: Hammond, M. {2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283-302. Webster-Stratton, C. H., Reid, M. Beauchaine, T. (2011). Combining parent and child training for voung children with ADHD. Journal of Clinical Child 8: Adolescent 40l2), 191-203. Webster-Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible vears teacher and child training programs in high-risk schools. Journal of child and 49{5), 471-483. Widome, R., Sieving, R. E., Harpin, S. A, Hearst, M. 0. {2008). Measuring neighborhood connection and the association with violence in young adolescents. lAdolesc Health, 43(5), 482-489. Wilcox, H. C., Kellam, S. 6., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Anthony, J. C. [2008). The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560??3. Wilson, W. J. {2011). When work disappears: The world of the new urban poor: Vintage. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. Wyman, P. A, Brown, C. H., LoMurrav, M., Schmeelk?Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 16534661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, Wu, K., 81 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 49 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Other Risk] Protective Factors for Suicide Suicide Attempts or ldeation Lead Sectors1 Government (local, Unemployment benefit programs state, Federal) Strengthen economic Other income supports Businessflabor . . . . . . Housmg pollcles Government {locaL The National Neighborhood Stabilization state, Federal) Program Coverage of mental health conditions in health insurance policies Health ca re Strengthen Mental Health Parity Laws 1/ access to Government (state, mental health Fadera? ca re Reducing access to lethal means among persons at-risk Government (local, intervening at suicide hot spots state) 1/ Safe storage practices Public Health Establish protective Organlzatlonal pollcles and culture Busmesstabor . Together for Life envnronments Government (local, US Air Force Suicide Prevention Program 1/ state, Fed eral) 50 Best Available Evidence Communitywbased policies to reduce excessive alcohol use Alcohol outlet density 1/ Government {local, state] Businessllabor Promote connectedness Peer norm approaches Public Health Sources of Strength Education Communitvrengagement activities Public Health Greening vacant urban spaces Government {local} Social emotional learning Public Health Youth Aware of Mental Health Program Education Signs of Suicide Teach co in 3 Good Behavior Game and problem- solvmg Sklils Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening Families 10-14 Gatekeeper training Public Health Mental Health First Aid Healthcare Screening combined with care management Healthcare Identify and 51 support people alt-risk Henry Ford Perfect Depression Core {Pre- cursor to Zero Suicide) Best Available Evidence Social Services Crisis Intervention Public Health Active foiio w?up con toct approaches EST for Suicide Prevention Notionoi Suicide Prevention Lifeiine Social Services Appiied Suicide intervention Training Treatment for people at risk of suicide Health ca re improving Mood Promoting Access to v? Social Services Coiioborotive Treatment Justice Coiiohorotive Assessment and Management v? Suic?doii CAMS Intervene to I lessen harms Dioiecticoi Behovioroi Therapy and prevent future risk Attachment?Based Fomiiv Therapy Treatment to prevent re-attempts Health ca re ED Brief intervention with Foiiow-up Visits Social Services v? Postvention Heaithcare 52 Best Available Evidence StandBy Response Service Safe messaging following a suicide Media Guidelines 1/ Public Health Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing the specific programmatic activities. Preventing Suicide: A Technical Package of Policy, Programs. and Practices Prepared by: Deb Stone, Set), MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical' Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hoary, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casev Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Christine Schuler National Institute for Occupational Safetv and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellvson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and deliverv of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessening harms and preventing future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention1 and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome.3 Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific wavs to advance the strategy. This can be accomplished through programs, policies, and practices. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberatelv results in injury or the potential for injurv.?1 Self-directed violence may be suicidal or non- suicio'oi in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fotoiself?directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs}!5 In 2015 (the most recent year of available death data), suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes.6 In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S.7 Overall suicide rates increased 28% from 2000 to 2015.6 Suicide is a problem throughout the life span; it is the third leading cause of death for youth 10-14 years of age, the second leading cause of death among people 15?24 and 25-34 years of age; the fourth leading cause among people 35 to 44 years of age, the fifth leading cause among people ages 45-54 and eighth leading cause among people 55-64 years of age.6 Suicide rates vary by race/ethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.IS Other population groups disproportionately impacted by suicide include middle-aged adults {whose rates increased 35% from 2000 to 2015, with steep increases seen among both males and females aged 35-64 years?; Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in decades)?; workers in certain occupational groups??1 and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority peers.12'14 Suicides reflect only a portion of the problem.15 Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.15 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide?vl? Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time}:5 The social ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk and protective factors identified in the literature.? Risk and protective factors for suicide exist at each level. For example, risk factors include: 0 Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness}:5 it is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide.13-19 Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status}:5 Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means}:5 These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts.??25 Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.? Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide The effects ofviolence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide.23'3? Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, 8 connectedness to one's community,31 school,32 family,33 caring and pro-social peers35 can enhance resilience and help reduce risk for suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities?f'f? In an early study, Crosby and Sacks?i1 estimated that 7% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death.? In a more recent study, in one state, Cerel et al?12 found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide andfor having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt.? Similarly, survivors of a loved one's suicide may experience ongoing pain and suffering including complicated grief,"M stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide.45r?15 Less discussed but no less important, are the financial and occupational effects on those left behind.? The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alone.? Adjusting for potential under-reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013.43 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.f3 The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable};5 While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach}!f9 Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public}5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on Specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context ifthe program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and conducting more evaluations may be necessary to address different population groups. The evidence? based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cutting Themes One important feature of the package is the complementary and potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact subsets of the population persons who have already made a suicide attempt). 10 Preventing Suicide Strategy Approach Strengthen economic supports I- Strengthen household financial security . Housing stabilization policies Strengthen access and Of suicide I Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas I Safer suicide care through systems change Create protective environments - Reduce access to lethal means among persons at- risk of suicide Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness 1 Peer norm programs I Community engagement activities Social-emotional learning programs Parenting skill and family relationship programs Gatekeeper training Crisis Intervention Treatment for people at-risk of suicide Treatment to prevent re?attempts Postvention i Safe reporting and messaging about suicide Teach coping and problem?solving skills Identify and support people at?risk Lessen harms and prevent future risk It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social-emotional learning programs, an approach under the Teach Coping and Problem- Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden ofsuicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, race/ethnicity, age, occupation, and other important population characteristics.5-5? Further, certain transition periods are also associated 11 with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian statusl?gr51 In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced.52 Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience.1 Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work.?55 These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts}. The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector Involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working? age individuals 25 to 64 years old.5557 Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation ofsuch financial stress may increase an individual's riskfor suicide or may indirectly increase risk by exacerbating related physical and mental health problems.58 Buffering these risks can, therefore, potentially protect against suicide. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates 13 I Reductions in eviction rates I Reductions in emotional distress - Reductions in rates of suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federai?Stote Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss.59 An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits {mean level: $73990 per person in US. constant dollars). The effects of unemployment benefit programs were also consistent by sex and age group.59 Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment {less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk.? Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliffe"1 examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin and Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied.El At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year.61 Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. 14 Housing stabilization policies. The Neighborhood Stabilization Program52 was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults.63 Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began] to 2010 (after it had peaked)?? Most of these suicides occurred prior to the actual loss ofthe decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide?!19 and the level of risk conferred by different types of mental illness previous research indicates that mental illness is an important risk factor for suicide?!? State-level suicide rates have also been found to be correlated with general mental health measures such as depressionFE?EQ Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions.m Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services."1 Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention.5 Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care.? Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: 0 Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity)?3 Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatient/outpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. If a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health providers; this shortage is particularly severe among low-income urban and rural communities.M There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and 16 federal programs loan repayment programs) and expanding the reach of health services through telephone, video and web-based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People Alt-Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes I Increased use of mental health services I Lower rates of treatment attrition I Reductions in depressive I Reductions in rates ofsuicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey on Drug Use and Health (NSDUH) is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and service utilization.SD Using data from this survey, Harris, Carpenter, and Baol? found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang 55' examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year.""9 Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps (NHSC), which offers 17 financial incentives to attract mental/behavioral health clinicians to underserved areas.? Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas (HPSAs) in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to the Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Teiementoi Heoith services refer to the use of telephone, video and web-based technologies for providing or care at a distance?9 TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities) to treat a wide range of mental health conditions. it can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide.? Further, Mohr and colleagues30 conducted a meta?analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone- administered compared to patients receiving face-to-face therapy.SD Th us, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered Perfect Depression Care,81 the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide.in An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 (110.3 to 47.6 per 100,000 population; p<.04] with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members 18 who accessed onlyr general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 19 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes.Em Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide?? For example, rates of suicide are high among middle- aged adults who comprise 42.6% ofthe workforce35; among certain occupational groupsmv?; and among people in detention facilities jail, prison),35 to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not.37'33 Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition.39?9? Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. - Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 35% of people who use a firearm in a suicide attempt will die from the injury).95 Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes,9537 and 2) people tend not to substitute a different method when a highly lethal method is unavailable or difficult to accessFB?gg Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: intervening at Suicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek helpm? 20 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms locked in a secure place in a gun safe or lock box), unloaded and separate from the ammunition; and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential settings}. Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling}, and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors (eg. depression, social Community?based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age.105 These policies are important because acute alcohol use has been found to be associated with more than one?third of suicides and approximately 40% of suicide Potential Outcomes Increases in safe storage of lethal means Reductions in rates of suicide Reductions in suicide attempts Increases in help-seeking Reductions in alcohol-related suicide deaths Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. 21 Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicidem?im For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year.103 Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to other jumping sites.108 Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the remavai of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from trauma centers, researchers found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents.110 Further, a recent systematic review of clinic and community?based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own.101 Another program, the Emergency Department Counseling an Access to Lethalr Means CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al89 found that at post-test 76% {of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le, all (100%) reported guns were currently locked up at post-test.89 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking.111 Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 22 Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement}. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problems and instead sees them as larger service- wide problems impacting the whole community.112 Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicidemThe program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch] found significantly lower rates of suicide after the program was launched than before.? These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years.? Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff; standardized intake screening and risk assessment; provision of shared information between staff members {especially in transitioning or transferring of inmates); varying levels of observation; safe physical environment; emergency response protocols; notification of suicidal behavior/suicide through the chain of command; and critical incident stress debriefing and death review can potentially reduce suicide.102 When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year.113 Other similar programs have seen declines in suicide both in the United States and internationally.1M Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density and risk factors for suicide, such as interpersonal violence and social suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that greater density of bars, specifically, is related to greater suicide and suicide attempts, particularly in rural areas.119 23 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, lack of connectedness, are among the chief causes for suicidality.120 Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others.121 Social connections can be formed within and between multiple levels of the social ecology,? for instance between individuals peers, neighbors, co-workers}, families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizationslm123 Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement), and improved mental health.12??v?5 Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole.121 Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. 0 Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings.126 0 Community engagement activities. Community engagement is an aspect of social capital.? Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. 24 Potential Outcomes I Increases in healthy coping attitudes and behaviors I Increases in referrals for youth in distress - Increases in help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. I Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (E- metropolitan, 12 rural), researchers found that the program improved adaptive norms regarding suicide, connectedness to adults, and school engagement.3E Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaclers??5 - Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other cross-cutting benefits, including reductions in firearm assaults and vandalism.125L129 25 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors.126 Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,13? surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness} characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters.131 Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social-emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem?solving skills. . Social-emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide.126 These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work.132 I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities.132 Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 26 Potential Outcomes I Reductions in suicide ideation Reductions in suicide attempts - Reductions in suicide risk behaviors depression, anxiety, conduct problems, substance abuse) I Improvements in help-seeking behavior - Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social-emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse.133 Social-emotional learning programs. The Youth Aware of Mental Health Program (MM) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problemsm In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, ?ve attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group, relative risk fell by 49.5%.134 Another example is the Good Behavior Game (656), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior.135 Two cohorts of youths participated in the program in 1985-85 and 1936-87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts overtime. With respect to distal suicide-related outcomes, an outcome evaluation ofthe 6136 indicated that individuals in the first cohort who were assigned to participate in (536 when 27 they were in the first grade reported half the adjusted odds of suicidal ideation and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect ofthe program was consistent for suicidal ideation regardless of whether baseline covariates were included.?5 The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of 656 students, neither suicidal ideation nor suicide attempts were significantly different between 686 and the control interventions.135 The researchers believed this may have been due to a lack of implementation fidelity, including less mentoring and monitoring of teachers. (EEG was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the first cohort of students. Results for the second cohort were generally smaller but in the desired direction.?5 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredibie Years is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence (all protective factors for suicide).132 The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the if? program on reducing internalizing such as anxiety and depression, and child conduct problems.131133 The program is also associated with improved problem-solving and conflict management; these skills were maintained at Additionally, the program demonstrated greater benefits in mother?rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.132 Additionally, Strengthening 10?14 is a program that involves sessions for parents, youth, and families with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths' interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide.?2 Strengthening Families has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use 28 among youth participants, as well as reduce depression, aicehol use, and drug use among participating families?? 29 Identify and Support People At-Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority Supporting people at?risk requires proactive case finding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible Internet-based services when appropriate] and engaging people in evidence-based care through such measures as collaborative treatment), remain key Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services.m Approaches The following approaches focus on identifying and SUpporting people at increased risk of suicide. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support peopie at risk?? a Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization?? Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Treatment for people at?risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with problem-solving and emotion regulation. Treatment usually takes place in a one-on-one or 30 group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative lie, between patient and therapist or care manager) and/or integrated care linkage between primary care and behavioral health care) can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk.1f?i?15D Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts.151 Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%~25% reattempt within a year, and of attempt survivors die by suicide within 1 to 5 years of their initial attempt.151 Potential Outcomes I Reductions in suicidal ideation - Reductions in suicide attempts I Reductions in suicide rates I Reductions in depression and feelings of hopelessness I Reductions in reattempts I Improvements in coping skills 0 Increases in treatment engagement and compliance with medications Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Skills Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: I-ileinman152 evaluated the training across the Notional Suicide Prevention Lifeline network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, the researchers found that callers who spoke with ASIST-trained counselors were significantly more likely to feel less 31 depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call, compared to callers who spoke to counselors. Counselors trained in ASiSTwere also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not result in more comprehensive suicide risk assessments than usual care training.152 Gatekeeper training has also been a primary component of the Garret Lee Smith Suicide Prevention Program, which has been funded in 50 states and 50 tribes. A multi?site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10-24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation.153 This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities (4.9 fewer attempts per 1000 youths).154 More than 29,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Nationai Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 ofthose completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1) to follow-up {time 3).155 Between time 2 (end of the call) to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.?5 Treatment for peeple at?risk of suicide. The improving Mood Promoting Access to Coiiaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase) by a depression care manager.155 The program has been shown to significantly improve quality of life, 32 and to reduce functional impairment, depression and suicidal ideation over 24-months of follow- up155-15? relative to patients who received care as usual. Collaborative Assessment and Management of Suicidaiity (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,WI in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual (intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed} found better treatment retention among the CAMS group and significant improvements in suicidal id eation, overall distress, and feelings of hopelessness at the 12 month follow-up.153 Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined.159 ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety.?El A randomized controlled trial ofABFTfound that suicidal adolescents assigned to ABFTexperienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%) and at 24 weeks (82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Soiutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows-up with both patients and providers 33 between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 20% of primary care patients.?1 also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively.161 Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 5, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively).162 Another example of treatment to prevent re-attempts involves active follow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation.151 These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years). In a meta-analysis conducted by Inagaki et al151 interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated.151 Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow?up approaches on death by suicide. In a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto, Bostrom 153 found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other 34 studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and Finallv; Cognitive Behavior Therapy for Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction; relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems; school or work-related difficulties) leading up to and following the suicidal event; safetv plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50%- reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usualF'E?E 35 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world.5 Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide?? Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion.153?159 Approaches Some approaches that can be used to lessen harms and reduce future risk of suicide include postvention and safe reporting and messaging following a suicide. I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief.1m I Safe reporting and messaging about suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. REports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotlinel, and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion.m Potential Outcomes I Reductions in suicidal ideation I Reductions in suicide attempts I Reductions in rates ofsuicide I Reductions in distress I Improvements in reporting following suicide I Reductions in contagion effects related to suicide 36 Evidence Current evidence suggests that postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postvention. One example of a postvention program with evidence of impact on risk and protective factors for suicide is the StondBy Response Service (Stond?y). StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs.172 In a study by Visser, Comans, and Scuffham,H2 StondBv clients were significantly less likely to be at high risk for suicidality (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program {48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendations for Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annualliyfi?9 Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide.1M Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery? of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reportsm 37 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the NotionolStrotegyfor Suicide Prevention,1 the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact ofsuicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Cure. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Sicilis to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 38 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can serve in an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 39 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical and/or medico-legal standards.4 uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems.? Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of-death data from death certificates? is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches.?5 Data from state and local Child Death Review teamsm and Suicide Death Review Teams {which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department).5 in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth 40 have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdosem The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies.W The Notional Survey on Drug Use and Health is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk.5'0 It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, 41 like many public health problems, suicide is preventable}5 and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide the release of the first world report on suicide? and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 42 References 10. 11. 12. 13. 14. 15. 16. 17. 13. 19. 20. US. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: 2012. National Action Alliance for Suicide Prevention. Action Alliance Priorities. 2017,- Frieden TR. Six components necessary for effective public health program implementation. Am Public Health. Crosby AE, Ortega L, Melanson C. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press; 2014. Centers for Disease Control and Prevention. Web-Based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Available online: Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. KE, Canham-Chervak M, Black SA, Dada EO, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2007?8. Prev. Lineberry TW, O'Connor SS. Suicide in the US Army. Mayo Clinic Proceedings. McIntosh WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholow B. Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep. Han B, Crosby AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?54year5 in the United States. Compr Kann L, McManus T, Harris WA, et al. Youth Risk Behavior Surveillance United States, 2015. MMWR CDC Surveill Summ. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: Evidence from a national study. Am Public Health. Stone DM, Luo F, Ouyang L, Lippy C, Hertz MF, Crosby AE. Sexual orientation and suicide ideation, plans, attempts, and medically serious attempts: evidence from local Youth Risk Behavior Surveys, 2001?2009. Am Public Health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer J. Suicidal thoughts and behaviors among adults aged >l=18 years?United States, 2008-2009. MMWR CDCSurveill Summ. Lipari R, Piscopo K, Kroutil LA, Kilmer Miller G. Suicidal thoughts and behavior among adults: Results from the 2014 National Survey on Drug Use and Health. NSDUH Data Review 2015; Dahlberg LL, Krug EG. Violence?a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?21. Owens 0. Fatal and non-fatal repetition of self~harm: Systematic review. Bri Olfson M, Gerhard T, Huang C, S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Bossarte RM, Karras E, Lu N, et al. Associations between the Dartment of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Health Rep (Washington, D.C.: 1974). 43 21. 22. 23. 24. 25. 26. 27. 23. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood. JAffect Disord. Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med. Klomek AB, Sourander A, Gould M. The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. Can Leeb RT, Lewis T, Zolotor AJ. A review of physical and mental health consequences ofchild abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. World Health Organization. Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence. Geneva: World Health Organization; 2013. Bellis IVIA, Hughes K, Leckenby N, et al. Adverse childhood experiences and associations with health- harming behaviours in young adults: surveys in eight eastern EurOpean countries. Bull World Health Organ. Haegerich TM, Dahlberg LL. Violence as a public health risk. American Journal of Lifestyle Medicine. Wilkins N, Tsao B, Hertz M, Davis R, Klevens J. Connecting the dots: An overview of the links among multiple forms of violence. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2014. Hamby S, J. The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociology. New York, NY: Springer; 2013. Kleiman EM, Riskind JH, Schaefer KE, Weingarden H. The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis. Carter M, McGee R, Taylor B, Williams S. Health outcomes in adolescence: associations with family, friends and school engagement. .lAdolesc. Maimon D, Browning CR, Brooks-Gunn J. Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse. Losel F, Fa rrington DP. Direct protective and buffering protective factors in the development of youth violence. Am Prev Med. 2012:4342 Suppl Wyman PA, Brown CH, LoMurray M, et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health. Dunne EJ, McIntosh JL, Dunne-Maxim K, eds. Suicide and its aftermath: Understanding and counseling the survivors. New York: Norton; 1987. Mishara BL. The impact of Suicide. New York: Springer; 1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. :Author; 2015. 44 41. 42. 43. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57'. 58. 59. 60. 51. 62. 63. Crosby AE, Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Lh?e Threat Behav. Cerel J, Maple M, De Venne A, Moore M, Flaherty C, Brown M. Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Heaith Rep (Washington, D.C.: 1974). Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav. Mitchell AM, Kim Y, Prigerson HG, Mortimer-Stephens M. Complicated grief in survivors ofsuicide. Crisis. Sudak H, Maxim K, Carpenter M. Suicide and stigma: a review of the literature and personal reflections. Acad Cerel J, McIntosh JL, Neimeyer RA, Maple M, Marshall D. The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide Life Threat Behav. Florence C, Simon T, Haegerich T, Luo F, Zhou C. Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR Moria Mortai Wkly Rep. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr., Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav. Silverman MM, Maris The prevention of suicidal behaviors: An overview. Suicide Life Threat Behav. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016: Desai RA, Dausey DJ, Rosenheck RA. Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am Turecki G. Epigenetics and suicidal behavior research pathways. Am Prev Med. Edwards RW, JumpervThurman P, Plested BA, Getting ER, Swanson L. Community readiness: Research to practice. Jaurnai of Community Hawkins JD, Catalano RF, Kuklinski MR. Communities that care. Encyciopedia of Criminoiagy and Criminai Justice: Springer; 2014:393-408. Plested BA, Edwards RW, Jumper?Thurman P. Community readiness: A handbook for successful change. Fort O: Tri-Ethnic Center for Prevention Research. 2006. Luo F, Florence CS, QuispeeAgnoli M, Duyang L, Crosby AE. Impact of business cycles on US suicide rates, 1928?2007. Am Public Health. Fowler KA, Gladden RM, Vagi KJ, Barnes J, Frazier L. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005?2010.Am Pubiic Heaith. Stack S, Wasserman l. Economic strain and suicide risk: a qualitative analysis. Suicide Life Threat Behav. Cylus J, (Slymour MM, Avendano M. Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am .i Epidemioi. Classen TJ, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ. 2012:21i3lz338-350. Flavin P, Radcliff B. Public policies and suicide rates in the American states. Sociai indicators Research. US. Department of Housing and Urban Development. Neighborhood Stabilization Program. 2017. Houle JN, Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith. 45 64. 65. 66. 67. 58. 69. 7D. 71. 72. 73. 74. 75. 76. 77. 73. 79. 80. 81. 82. 83. 84. 85. 86. 87. ArsenauIt-Lapierre G, Kim C, Turecki (3. diagnoses in 3275 suicides: a meta-analysis. BMC 2004;437. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-a nalysis. Tyrer P, Reed GM, Crawford MJ. Classification, assessment, prevalence, and effect of personality disorder. Lancet. Harris EC, Barraclough B. Excess mortality of mental disorder. Mark TL, Shern DL, Bagalman JE, Cao Z. Ranking America?s mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 2007. Lang M. The impact of mental health insurance laws on state suicide rates. Health Econ. 88. Wang PS, Demler O, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Am Public Health. Cunningham PJ. Beyond parity: primary care physicians' perspectives on access to mental health care. Health Aff (Millwood). Coffey CE. Building a system of perfect depression care in behavioral health. it Comm Qual Patient Saf. The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 (MHPEA), HR 1424, 110th Congress. 2008. U.S. Department of Health and Human Services Health Resources and Services Administrations. Designated Health Professional Shortage Areas Statistics, As of November 4, 2016. 2016. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author; 2014. Harris KM, Carpenter C, Bao Y. The effects of state parity laws on the use of mental health care. Med Care. 2006;44l6lz499?505. Health Resources Services Administration. National Health Service Corps. 201?; U.S. Department of Health and Human Services Health Resources and Services Administrations. National Health Service Corp Clinician Retention: A Story of Dedication and Commitment. 2016. Hailey D, Raine R, Ohinmaa A. The effectiveness of telemental health applications: a review. The Can Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect of telephone~administered on of depression and attrition: A meta-a nalysis. Clinical Science and Practice. 2008;15l3lz243-253. Coffey CE. Pursuing perfect depression care. Serv. Coffey CE, Coffey Ml, Ahmedani BK. An update on perfect depression care. Serv. Coffey M, Coffey C, Ah medani BK. Suicide in a health maintenance organization population. Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep. Toosi M. Labor force projections to 2024: The labor force is growing, but slowly. Washington, D.C.: Bureau of Labor Statistics; Labor Review. 2015:1-33. Noonan ME. Mortality in State Prisons, 2001-2014 Bureau oflustice Statistical Tables. 2016;250150lDecember). Knox KL, Pflanz S, Talcott (3W, et al. The US Air Force suicide prevention program: implications for public health policy. Am Public Health. 46 38. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 103. 109. 110. National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Blaeprintfor Workplace Suicide Prevention. Washington, D.C. Author; 2015. Ru nyan CW, Becker A, Brandspigel S, Barber C, Trudeau A, Novins D. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJr Emerg Med. Miller M, Warren M, Hemenway D, Azrael D. Firearms and suicide in US cities. in} Prev. Crosby AE, Espitia-Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Alcohol: Science, Policy and Public Health. 2013:190-193. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: 3 gender-stratified analysis of the National Violent Death Reporting System. in} Prev. Beautrais AL, Gibb SJ, Fergusson DM, Horwood LJ, La rkin GL. Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Journal of Correctional Health Care. 242. Elnour AA, Harrison J. Lethality of suicide methods. in} Prev. Simon OR, Swarm AC, Powell KE, Potter LB, Kresnow MJ, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behov. 2001;32{1 Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin Hawton K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. Yip P, Caine E, Yousuf 5, Chang S-S, Wu K, Chen Means restriction for suicide prevention. Lancet. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic raview. BMC Public Health. Rowhani-Rahbar A, Simonetti 1A, Rivara FP. Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev. Hayes LM. Suicide prevention in correctional facilities: reflections and next steps. int Law Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction. Escobedo LG, Ortiz M. The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Xuan Z, Naimi TS, Kaplan MS, et al. Alcohol Policies and Suicide: A Review of the Literature. Alcohol Clin Exp Res. Cherpitel CJ, Borges GLG, Wilcox HC. Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research. 2004;28l5 Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: A systematic review and meta-analysis. Lancet Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ Grossman DC, Mueller BA, Riedy C, et al. Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA. 47 111. 112. 113. 114. 115. 115. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 12?. 128. 129. 130. 131. 132. 133. Mishara BL, Martin N. Effects of a comprehensive police suicide prevention program. Crisis. Knox KL, Litts DA, Talcott (3W, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. Hayes LM. Prison Suicide: An Overview and a Guide to Prevention. The Prison Journal. 456. Barker E, Kolves K, De Leo D. Management ofSuicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research. 20143881222240. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol-related damage. I. The distribution of consumption model. 1 Stud Alcohol. Gruenewald PJ, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res. Lipton R, Gruenewald P. The spatial dynamics of violence and alcohol outlets. .lStud Alcohol. Lippy C, DeGue 5. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Violence, Abuse. Johnson FW, Gruenewald Pl, Remer LG. Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res. Durkheim E. Suicide: A study in sociology Spaulding G. Simpson, trans). Glencoe, it: Free work published 189?). 1897;1951. Centers for Disease Control and Prevention. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. 2009; Available at: a.pdf. Muennig P, Cohen AK, Palmer A, Zhu W. The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med. Beyer KM, Layde PM, Hamberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women. Whitley R, McKenzie K. Social capital and review of the literature. Harv Rev De Silva MJ, McKenzie K, Harpham T, Huttly SR. Social capital and mental illness: a systematic review. Epidemiol Community Health. Wyman PA. Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med. Suppl Centers for Disease Control and Prevention. Principles of community engagement. TSDR Committee on Community Engagement. 199?. Branas CC, Cheney RA, MacDonald JM, Tam VW, Jackson TD, Ten Have TR. A difference-in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol. Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald JM. Urban blight remediation as a cost-beneficial solution to firearm violence. Am Public Health. 2016;106l12lz21SB-2164. Ba ndura A. Socialfoundations of thought and action: A social cognitive theory. Prentice-Hall, Inc; 1986. Pollock LR, Williams JM. Problem-solving in suicide attempters. Med. Herman KC, Borden LA, Reinke WM, Webster-Stratton C. The Impact of the Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q. Knox MS, Bu rkhart K, Hunter KE. ACT against violence parents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. 2010. 48 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. Wasserman D, Hoven CW, Wasserman C, et al. School-based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. Lancet. Wilcox HC, Kellam 56, Brown CH, et al. The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend. 2008;95 Suppl 1:560-73. Kellam 56, Brown CH, Pod uska JM, et al. Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend. 2008;95 Suppl 1:55-528. Webster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation ofthe incredible years teacher and child training programs in high-risk schools. Journal of Child and Webster-Stratton CH, Reid MJ, Beauchaine T. Combining parent and child training for young children with ADHD. Journal of Clinical Child 82 Adolescent Reid MJ, Webster-Stratton C, Hammond M. Follow-up ofchildren who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy. Webster-Stratton C, Hammond M. Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin Webster?Stratton C, Reid MJ, Hammond M. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child Spoth RL, Guyll M, Day 5X. Universal family-focused interventions in alcohol?use disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics; 2016. Jobes BA. The Collaborative Assessment and Management ofSuicidality (CAMS): an evolving evidence based clinical approach to Suicidal risk. Suicide Life Threat Behav. Wilcox HC, Wyman PA. Suicide Prevention Strategies for improving Population Health. Child Adolesc Clin Isaac M, Elias B, Kat: LY, et al. Gatekeeper training as a preventative intervention for suicide: a systematic review. Can Gould MS, Munfakh JL, Kleinman M, Lake AM. National suicide prevention lifeline: enhancing mental health care for suicidal individuals and other people in crisis. Suicide Life Threat Behav. 35. Gilbody S, Bower P, Fletcher J, Richards D, Sutton AJ. Collaborative care for depression: a cumulative meta?analysis and review of longer?term outcomes. Arch int Med. 14?232 1. Archer i, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews. 2012;10lArt No Bruce ML, Ten Have TR, Reynolds CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA. Inagaki M, Kawashima Kawanishi C, et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Affect Disord. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. Walrath C, Garraza LG, Reid H, Goldston DB, McKeon R. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. 49 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 1149. Gould MS, Kalafat J, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. Unutzer J, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc. Comtois KA, Jobes DA, 5. O'Connor S, et al. Collaborative assessment and management of suicidality feasibility trial for next-day appointment services. Depress Anxiety. Linehan MM, Comtois KA, Murray AM, et al. Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Diamond GS, Wintersteen MB, Brown GK, et al. Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. JArn Acad Chiia?Aa?aiesc 2010;49i2ileZ-131. Ru benstein LV, Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Systems, Heaith. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buii World Heaith Organ. Motto JA, Bostrom AG. A randomized controlled trial of postcrisis suicide prevention. Serv. Hassanian?Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self~poisoning. Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3?Month Intervention. Suicide Life Threat Behav. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects ofsuicide bereavement on mental health and suicide risk. Lancet Etzersdorfer E, Sonneck G. Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res. Niederkrotenthaler T, Sonneck G. Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21 Szumilas M, Kutcher S. Post-suicide intervention programs: A systematic review. Can Pubiic Heaith. Bohanna I, Wang X. Media guidelines for the responsible reporting of suicide: a review of effectiveness. Crisis. Visser VS, Comans TA, Scuffham PA. Evaluation Of The Effectiveness Community-Based Crisis Intervention Program For People Bereaved By Suicide. Jaurnai of Community 28. Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav. 50 174. 175. 176. 177. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. 2010;341:c5841. Centers for Disease Control and Prevention. National Violent Death Reporting Svstem. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2017. Available online: The National Center for the Review Prevention of Child Deaths. U.S. Child Death Review Programs. Centers for Disease Control and Prevention, Brener ND, Karin L, et al. Methodology of the Youth Risk Behavior Surveillance System-?013. MMWR Recomm Rep. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment bene?t programs 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, 1 Neighborhood Stabilization Program a era Coverage of mental health conditions in health insurance policies Mental Health Parity Laws 1/ 1/ Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps suicide care Healthcare Telemental Health (TMH) Social services Safer suncide care through systems change Henry Ford Pen?ect Depression Care (Pres V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt Government (local . . . intervening at hot spots state) Create - Safe storage practices v/ - protective Public Health environments Emergency Department Counseling on Healthcare 52 Organizational policies and culture Best Available Evidence Together for Life 1/ US Air Force Suicide Prevention Program v? Correctional suicide prevention 1/ BusinessiLabor Jus?ce Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Government {locaL state) Promote cannectedness Alcohol outlet density Business/labor Peer norm programs Public Health Sources ofStrength Education Community engagement activities Public Health Greening vacant urban spaces Government {local} Teach coping and problem- solving skills Social-emotional learning programs Public Health Youth Aware of Mental Health Program Education Good Behavior Gome Parenting skill and family relationship approaches The incredible Years Public Health Frlur'etinn 53 Strengthening Families 10?14 Best Available Evidence Identify and at-risk Gatekeeper training support people Public Health Applied Suicide intervention Skills Training Healthcare Garret Lee Smith Federal Grant Program 1/ v? Crisis Intervention Public Health . A . . . Social Services National Prevention Lifeline 1/ Treatment for people at risk of suicide improving Mood Promoting Access to f, Collaborative Treatment Collaborative Assessment and Management Health ca re of5uicidality (CAMS) Dialectical Behavioral Therapy (DBT) v? ervlces Justice Attachment-Based Family Therapy (ABFT) 1/ Translating initiatives for Depression into E??ective Solutions project Treatment to prevent re~attempts ED Brief intervention with Follow-up Visits Health to re Active fallow?up contact approaches v" Social services EST for Suicide Prevention 1/ 54 Best Available Evidence Postvention Lassen harms Stond?y Response Service 1/ 1/ ea care and prevent future risk Safe reporting and message about suicide Public Health 1/ Media Gurdeimes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. Preventing Suicide: A Technical Package of Policy, Programs. and Practices Prepared by: Deb Stone, Set), MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical' Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hoary, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casev Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Christine Schuler National Institute for Occupational Safetv and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellvson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and deliverv of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessening harms and preventing future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention1 and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome.3 Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific wavs to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injurv.?El Self-directed violence may be suicidoi or non- suicidai in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fatolself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs}!5 In 2015 (the most recent year of available death data), suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes.6 In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S.7 Overall suicide rates increased 28% from 2000 to 2015.6 Suicide is a problem throughout the life span; it is the third leading cause of death for youth 10-14, the second leading cause of death among people 15?24 and 25-34 years; the fourth leading cause among people 35 to 44, the fifth leading cause among people 45-54 and eighth leading cause among people 55-64.6 Suicide rates vary by race/ethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.l5 Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 33% from 2000 to 2015, with steep increases seen among both males and females aged 35-64 years?; Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in workers in certain occupational groups?lr11 and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority peers.?M Suicides reflect only a portion of the problem-l5 Substantially more people are hospitalized as a result of nonfatal suicidal behavior lie. suicide attempts} than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.15 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicidefir1E5 Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time.? The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature.? Risk and protective factors for suicide exist at each level. For example, risk factors include: 0 Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress Community level: inadequate community connectedness, barriers to health care ie.g., lack of access to providers and medications) Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness}:5 it is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide?!19 Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status}!5 Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means.1-5 These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.? Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide attempts?:27 The effects ofviolence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress - factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide.?30 Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s communityf?1 school?"2 family,33 caring adultsf"435 and pro-social peers35 can enhance resilience and help reduce risk for suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities?l?f? In an early study, Crosby and Sacks41 estimated that ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death.41 in a more recent study, in one state, Cerel et al42 found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attemptii3 Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief,? stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide.""5"lI3 Less discussed but no less important, are the financial and occupational effects on those left behind.? The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alone.? Adjusting for potential under?reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013.43 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.? The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable?5 While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach}!49 Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public}!5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or 9 systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; cl meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U5. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and conducting more evaluations may be necessary to address different population groups. The evidence- based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cu [ting Themes One important feature of the package is the complementary and potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below]. The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact subsets ofthe populations persons who have already made a suicide attempt). 10 Preventing Suicide Strategy Approach Strengthen economic supports - Strengthen household financial security a Housing stabilization policies Strengthen access and delivery of suicide I Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas I Safer suicide care through systems change Create protective environments 0 Reduce access to lethal means among persons at? risk of suicide I Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness Peer norm programs In Community engagement activities coping and problem-solving Skills - Social-emotional learning programs Parenting skill and family relationship programs Identify and sopport people at-risk Gatekeeper training I Crisis Intervention I Treatment for people at-risk of suicide - Treatment to prevent re?attempts Postvention Ir Safe reporting and messaging about suicide Lessen harms and prevent future risk it is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teoch Coping and Problem- Soiving strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics.5-5G Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting 11 their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced.52 Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience.1 Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work.53?55 These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector Involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working? age individuals 25 to 64 years old.5557 Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation ofsuch financial stress may increase an individual's riskfor suicide or may indirectly increase risk by exacerbating related physical and mental health problems.58 Buffering these risks can, therefore, potentially protect against suicide. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates 13 I Reductions in eviction rates I Reductions in emotional distress - Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federai?Stote Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss.59 An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits {mean level: $73990 per person in US. constant dollars). The effects of unemployment benefit programs were also consistent by sex and age group.59 Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment {less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk.? Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliffe"1 examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin and Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied.El At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year.61 Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. 14 Housing stabilization policies. The Neighborhood Stabilization Program52 was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults.63 Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began] to 2010 (after it had peaked)?? Most of these suicides occurred prior to the actual loss ofthe decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide?!19 and the level of risk conferred by different types of mental illness previous research indicates that mental illness is an important risk factor for suicide?!? State-level suicide rates have also been found to be correlated with general mental health measures such as depressionFE?EQ Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions.m Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services."1 Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention.5 Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care.? Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: 0 Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity)?3 Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatient/outpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. If a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health providers; this shortage is particularly severe among low-income urban and rural communities.M There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and 16 federal programs loan repayment programs} and expanding the reach of health services through telephone, video and web-based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People Alt-Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes I Increases in access to mental health services I Increase in utilization of mental health services I Reductions in of mental illnesses and suicidality I Reductions in rates ofsuicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey on Drug Use and Health (NSDUH) is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and service utilization?? Using data from this survey, Harris, Carpenter, and Baol? found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang 55' examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year.59 Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps (NHSC), which offers 17 financial incentives to attract mental/behavioral health clinicians to underserved areas.? Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas (HPSAs) in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to the Ali-i561?8 Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Teiementoi health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance.T9 TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities) to treat a wide range of mental health conditions. it can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide.? Further, Mohr and colleagues30 conducted a meta?analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone- administered compared to patients receiving face-to-face therapy.SD Th us, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered Perfect Depression Care,81 the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide.431 An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 829639?!32 Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 (110.3 to 47.6 per 100,000 population; p<.04] with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members 18 who accessed onlyr general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 19 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes.Em Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide?? For example, rates of suicide are high among middle- aged adults who comprise 42.6% ofthe workforce35; among certain occupational groupsmv?; and among people in detention facilities jail, prison),35 to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not.37'33 Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition.39?9? Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. - Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 35% of people who use a firearm in a suicide attempt will die from the injury).95 Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes,9537 and 2) people tend not to substitute a different method when a highly lethal method is unavailable or difficult to accessFB?gg Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: intervening at Suicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek helpm? 20 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms--locked in a secure place le.g., in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts?grm1 0 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential settings}. Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help}, skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors leg. depression, social isolationi??i?i102 I Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicideslm?m Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking again-5 These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide Potential Outcomes Increases in safe storage of lethal means I Reductions in rates of suicide I Reductions in suicide attempts I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths 21 Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspat interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicidem?lmi For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year.1mg Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to other jumping sites.103 Further evidence for the effectiveness of bridge barriers was demonstrated by a studyr examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 3? counties in Washington, Oregon, and Missouri, and from 5 trauma centers, researchers found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescentsm Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own.?31 Another program, the Emergency Department Counseling on Access to Lethal Means CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. in a pre-post quality improvement project, Runyan et al39 found that at post-test 76% (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test.89 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness 22 of existing resources. The program also included an education campaign to improve awareness and help-seeking.111 Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city 29.0 per 100,000; 111 Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community.112 Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide.112The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years postulaunch) found significantly lower rates of suicide after the program was launched than before.? These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years.87 Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behavior/suicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide?? When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following yearxl13 Other similar programs have seen declines in suicide both in the United States and internationally?1 Community?based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density and risk factors for suicide, such as interpersonal violence and social connectedness,1l5'113 suggest that measures to 23 reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that greater density of bars, specifically, is related to greater suicide and suicide attempts, particularly in rural areas.119 24 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality.12D Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others.121 Social connections can be formed within and between multiple levels ofthe social ecology,? for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizationslm123 Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement), and improved mental health.12??v?5 Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole.121 Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. 0 Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings.126 0 Community engagement activities. Community engagement is an aspect of social capital.? Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. 25 Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors - Increases in referrals for youth in distressed I Increases help-seeking behaviors 0 Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. I Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools {6 metropolitan, 12 rural), researchers found that the program improved adaptive norms regarding suicide, connectedness to adults, and school engagement.3H5 Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history ofsuicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders.3'E I Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other cross-cutting benefits, including reductions in firearm assaults and vandalismm'lzg 26 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors.126 Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,13? surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness} characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters.131 Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem?solving skills. . Social-emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide.126 These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work.132 I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities.132 Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 27 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors li.e., depression, anxiety, conduct problems, substance abuse) - Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abusem Social?emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems.134 In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to mm were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group, relative risk fell by 49.6%.134 Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior.135 Two cohorts of youths participated in the program in 1985-86 and 1986-87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the 1336 indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation and 28 suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included.?5 The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between (586 and the control interventions.135 The researchers believed this may have been due to a lack of implementation fidelity, including less mentoring and monitoring of teachers. 636 was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the first cohort of students. Results for the second cohort were generally smaller but in the desired direction.135 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (in is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence (all protective factors for suicide}.132 The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the program on reducing internalizing such as anxiety and depression, and child conduct problems?i?m13B The program is also associated with improved problem-solving and conflict management; these skills were maintained at Additionally, the program demonstrated greater benefits in mother-rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.132 Additionally, Strengthening For Parents and Youth 10-14 is a program that involves sessions for parents, youth, and families with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide??- Strengthening Families has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families?? 29 Identify and Support People At-Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority Supporting people at?risk requires proactive case finding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible Internet-based services when appropriate] and engaging people in evidence-based care through such measures as collaborative treatment}, remain key chaIlenges.31r1?i?i?1?15 Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services.m Approaches The following approaches focus on identifying and SUpporting people at increased risk of suicide. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support peopie at risk?? a Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization?? Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Treatment for people at?risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with problem-solving and emotion regulation. Treatment usually takes place in a one-on-one or 30 group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/or integrated care linkage between primary care and behavioral health care) can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts.151 Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%~25% reattempt within a year, and of attempt survivors die by suicide within 1 to 5 years of their initial attempt.151 Potential Outcomes Reductions in suicide attempts Reductions in suicide rates Reductions in suicide deaths Reductions in of mental illnesses I Reductions in suicidal ideation I Reductions in mental health-related sequelae I Reductions in re-attempts I Increases in connectedness I Improvements in coping skills I Increases in identification of individuals at-risk for suicidal behavior I Increases in treatment engagement and compliance by at-risk individuals I Increases in community members trained to identify at-risk individuals I Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Training (ASIST) is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to 31 identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman152 evaluated the training across the Notionoi Suicide Prevention Lifeline network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, the researchers found that callers who spoke with ASIST?trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call, compared to callers who spoke to counselors. Counselors trained in ASiSTwere also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not result in more comprehensive suicide risk assessments than usual care training?:1 Gatekeeper training has also been a primary component of the Garret Lee Smith Suicide Prevention Program, which has been funded in 50 states and S0 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10-24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation.153 This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities {4.9 fewer attempts per 1000 youths).154 More than 79,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 ofthose completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1) to follow-up {time 3).155 Between time 2 (end of the call] to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.155 32 Treatment for people atnrisk of suicide. The improving Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase) by a depression care manager.155 The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow- up155'157' relative to patients who received care as usual. Collaborative Assessment and Management of Suicidaiity (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,144 in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual [intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal id eation, overall distress, and feelings of hopelessness at the 12 month follow-up.153 Other examples include Diaiectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DST were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment {23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined.159 ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety.150 A randomized controlled trial ofABFTfound that suicidal adolescents assigned to ABFTexperienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than 33 did adolescents receiving enhanced usual care {69.2% vs. 34.6%} and at 24 weeks (82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Solutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 70% of primary care patients.151 also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively.161 Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus Another example of treatment to prevent re-attempts involves active foiiow?up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation.151 These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years]. In a meta-analysis conducted by lnagaki et al151 interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated.151 Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. 34 in a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto, Bostrom 153 found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide id eation and attem pts.15?5?r165 Finally, Cognitive Behavior Theropyfor Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual.?5 35 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world.5 Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide?? Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion??a?l?g While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care for the bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions) and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. II Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief.1m I Safe reporting and messaging about suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline), and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion.m Potential Outcomes I Reductions in suicidal ideation Reductions in suicide attempts 36 I Reductions in distress I Increases in treatment seeking and engagement - Improvements in reporting following suicide 0 Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service (StondBy), provides clients with face?to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs?? In a study by Visser, Comans, and Scuffham,?2 StondBy clients were significantly less likely to be at high risk for suicidality (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendetionsfor Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually.1?59 Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide?M Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reportsm 37 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the NotionolStrotegyfor Suicide Prevention,1 the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact ofsuicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Cure. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Sicilis to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 38 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 39 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andlor medicolegal standards.?1 uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems.? Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of-death data from death certificates? is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches.?5 Data from state and local Child Death Review teamsm and Suicide Death Review Teams {which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department).5 in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth 40 have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdosem The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies.W The Notional Survey on Drug Use and Health is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk.5'0 It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, 41 like many public health problems, suicide is preventable}5 and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide the release of the first world report on suicide? and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 42 References 10. 11. 12. 13. 14. 15. 16. 17. 13. 19. 20. US. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: 2012. National Action Alliance for Suicide Prevention. Action Alliance Priorities. 2017,- Frieden TR. Six components necessary for effective public health program implementation. Am Public Health. Crosby AE, Ortega L, Melanson C. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press; 2014. Centers for Disease Control and Prevention. Web-Based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Available online: Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. KE, Canham-Chervak M, Black SA, Dada EO, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2007?8. Prev. Lineberry TW, O'Connor SS. Suicide in the US Army. Mayo Clinic Proceedings. McIntosh WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholow B. Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep. Han B, Crosby AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?54year5 in the United States. Compr Kann L, McManus T, Harris WA, et al. Youth Risk Behavior Surveillance United States, 2015. MMWR CDC Surveill Summ. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: Evidence from a national study. Am Public Health. Stone DM, Luo F, Ouyang L, Lippy C, Hertz MF, Crosby AE. Sexual orientation and suicide ideation, plans, attempts, and medically serious attempts: evidence from local Youth Risk Behavior Surveys, 2001?2009. Am Public Health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer J. Suicidal thoughts and behaviors among adults aged >l=18 years?United States, 2008-2009. MMWR CDCSurveill Summ. Lipari R, Piscopo K, Kroutil LA, Kilmer Miller G. Suicidal thoughts and behavior among adults: Results from the 2014 National Survey on Drug Use and Health. NSDUH Data Review 2015; Dahlberg LL, Krug EG. Violence?a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56. Owens 0. Fatal and non-fatal repetition of self~harm: Systematic review. Bri Olfson M, Gerhard T, Huang C, S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Bossarte RM, Karras E, Lu N, et al. Associations between the Dartment of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Health Rep (Washington, D. C. 1974). 43 21. 22. 23. 24. 25. 26. 27. 23. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood. JAffect Disord. Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med. Klomek AB, Sourander A, Gould M. The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. Can Leeb RT, Lewis T, Zolotor AJ. A review of physical and mental health consequences ofchild abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. World Health Organization. Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence. Geneva: World Health Organization; 2013. Bellis IVIA, Hughes K, Leckenby N, et al. Adverse childhood experiences and associations with health- harming behaviours in young adults: surveys in eight eastern EurOpean countries. Bull World Health Organ. Haegerich TM, Dahlberg LL. Violence as a public health risk. American Journal of Lifestyle Medicine. Wilkins N, Tsao B, Hertz M, Davis R, Klevens J. Connecting the dots: An overview of the links among multiple forms of violence. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2014. Hamby S, J. The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociology. New York, NY: Springer; 2013. Kleiman EM, Riskind JH, Schaefer KE, Weingarden H. The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis. Carter M, McGee R, Taylor B, Williams S. Health outcomes in adolescence: associations with family, friends and school engagement. .lAdolesc. Maimon D, Browning CR, Brooks-Gunn J. Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse. Losel F, Fa rrington DP. Direct protective and buffering protective factors in the development of youth violence. Am Prev Med. 2012:4342 Suppl Wyman PA, Brown CH, LoMurray M, et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health. Dunne EJ, McIntosh JL, Dunne-Maxim K, eds. Suicide and its aftermath: Understanding and counseling the survivors. New York: Norton; 1987. Mishara BL. The impact of Suicide. New York: Springer; 1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. :Author; 2015. 44 41. 42. 43. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57'. 58. 59. 60. 51. 62. 63. Crosby AE, Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Lh?e Threat Behav. Cerel J, Maple M, De Venne A, Moore M, Flaherty C, Brown M. Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Heaith Rep (Washington, D. C. 1974}. Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav. Mitchell AM, Kim Y, Prigerson HG, Mortimer-Stephens M. Complicated grief in survivors ofsuicide. Crisis. Sudak H, Maxim K, Carpenter M. Suicide and stigma: a review of the literature and personal reflections. Acad Cerel J, McIntosh JL, Neimeyer RA, Maple M, Marshall D. The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide Life Threat Behav. Florence C, Simon T, Haegerich T, Luo F, Zhou C. Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR Moria Mortai Wkly Rep. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr., Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav. Silverman MM, Maris The prevention of suicidal behaviors: An overview. Suicide Life Threat Behav. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016: Desai RA, Dausey DJ, Rosenheck RA. Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am Turecki G. Epigenetics and suicidal behavior research pathways. Am Prev Med. Edwards RW, JumpervThurman P, Plested BA, Getting ER, Swanson L. Community readiness: Research to practice. Jaurnai of Community Hawkins JD, Catalano RF, Kuklinski MR. Communities that care. Encyciopedia of Criminoiagy and Criminai Justice: Springer; 2014:393-408. Plested BA, Edwards RW, Jumper?Thurman P. Community readiness: A handbook for successful change. Fort O: Tri-Ethnic Center for Prevention Research. 2006. Luo F, Florence CS, QuispeeAgnoli M, Duyang L, Crosby AE. Impact of business cycles on US suicide rates, 1928?2007. Am Public Health. Fowler KA, Gladden RM, Vagi KJ, Barnes J, Frazier L. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005?2010.Am Pubiic Heaith. Stack S, Wasserman l. Economic strain and suicide risk: a qualitative analysis. Suicide Life Threat Behav. Cylus J, (Slymour MM, Avendano M. Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am .i Epidemioi. Classen TJ, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ. 2012:21i3lz338-350. Flavin P, Radcliff B. Public policies and suicide rates in the American states. Sociai indicators Research. US. Department of Housing and Urban Development. Neighborhood Stabilization Program. 2017. Houle JN, Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith. 45 64. 65. 66. 67. 58. 69. 7D. 71. 72. 73. 74. 75. 76. 77. 73. 79. 80. 81. 82. 83. 84. 85. 86. 87. ArsenauIt-Lapierre G, Kim C, Turecki (3. diagnoses in 3275 suicides: a meta-analysis. BMC 2004;437. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-a nalysis. Tyrer P, Reed GM, Crawford MJ. Classification, assessment, prevalence, and effect of personality disorder. Lancet. Harris EC, Barraclough B. Excess mortality of mental disorder. Mark TL, Shern DL, Bagalman JE, Cao Z. Ranking America?s mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 2007. Lang M. The impact of mental health insurance laws on state suicide rates. Health Econ. 88. Wang PS, Demler O, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Am Public Health. Cunningham PJ. Beyond parity: primary care physicians' perspectives on access to mental health care. Health Aff (Millwood). Coffey CE. Building a system of perfect depression care in behavioral health. it Comm Qual Patient Saf. The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 (MHPEA), HR 1424, 110th Congress. 2008. U.S. Department of Health and Human Services Health Resources and Services Administrations. Designated Health Professional Shortage Areas Statistics, As of November 4, 2016. 2016. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author; 2014. Harris KM, Carpenter C, Bao Y. The effects of state parity laws on the use of mental health care. Med Care. 2006;44l6lz499?505. Health Resources Services Administration. National Health Service Corps. 201?; U.S. Department of Health and Human Services Health Resources and Services Administrations. National Health Service Corp Clinician Retention: A Story of Dedication and Commitment. 2016. Hailey D, Raine R, Ohinmaa A. The effectiveness of telemental health applications: a review. The Can Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect of telephone~administered on of depression and attrition: A meta-a nalysis. Clinical Science and Practice. 2008;15l3lz243-253. Coffey CE. Pursuing perfect depression care. Serv. Coffey CE, Coffey Ml, Ahmedani BK. An update on perfect depression care. Serv. Coffey M, Coffey C, Ah medani BK. Suicide in a health maintenance organization population. Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep. Toosi M. Labor force projections to 2024: The labor force is growing, but slowly. Washington, D.C.: Bureau of Labor Statistics; Labor Review. 2015:1-33. Noonan ME. Mortality in State Prisons, 2001-2014 Bureau oflustice Statistical Tables. 2016;250150lDecember). Knox KL, Pflanz S, Talcott (3W, et al. The US Air Force suicide prevention program: implications for public health policy. Am Public Health. 46 38. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 103. 109. 110. National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Blaeprintfor Workplace Suicide Prevention. Washington, D.C. Author; 2015. Ru nyan CW, Becker A, Brandspigel S, Barber C, Trudeau A, Novins D. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJr Emerg Med. Miller M, Warren M, Hemenway D, Azrael D. Firearms and suicide in US cities. in} Prev. Crosby AE, Espitia-Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Alcohol: Science, Policy and Public Health. 2013:190-193. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: 3 gender-stratified analysis of the National Violent Death Reporting System. in} Prev. Beautrais AL, Gibb SJ, Fergusson DM, Horwood LJ, La rkin GL. Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Journal of Correctional Health Care. 242. Elnour AA, Harrison J. Lethality of suicide methods. in} Prev. Simon OR, Swarm AC, Powell KE, Potter LB, Kresnow MJ, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behov. 2001;32{1 Deisenhammer EA, Ing CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin Hawton K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. Yip P, Caine E, Yousuf 5, Chang S-S, Wu K, Chen Means restriction for suicide prevention. Lancet. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic raview. BMC Public Health. Rowhani-Rahbar A, Simonetti 1A, Rivara FP. Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev. Hayes LM. Suicide prevention in correctional facilities: reflections and next steps. int Law Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction. Escobedo LG, Ortiz M. The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Xuan Z, Naimi TS, Kaplan MS, et al. Alcohol Policies and Suicide: A Review of the Literature. Alcohol Clin Exp Res. Cherpitel CJ, Borges GLG, Wilcox HC. Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research. 2004;28l5 Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: A systematic review and meta-analysis. Lancet Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ Grossman DC, Mueller BA, Riedy C, et al. Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA. 47 111. 112. 113. 114. 115. 115. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 12?. 128. 129. 130. 131. 132. 133. Mishara BL, Martin N. Effects of a comprehensive police suicide prevention program. Crisis. Knox KL, Litts DA, Talcott (3W, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. Hayes LM. Prison Suicide: An Overview and a Guide to Prevention. The Prison Journal. 456. Barker E, Kolves K, De Leo D. Management ofSuicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research. 20143881222240. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol-related damage. I. The distribution of consumption model. 1 Stud Alcohol. Gruenewald PJ, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res. Lipton R, Gruenewald P. The spatial dynamics of violence and alcohol outlets. .lStud Alcohol. Lippy C, DeGue 5. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Violence, Abuse. Johnson FW, Gruenewald Pl, Remer LG. Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res. Durkheim E. Suicide: A study in sociology Spaulding G. Simpson, trans). Glencoe, it: Free work published 189?). 1897;1951. Centers for Disease Control and Prevention. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. 2009; Available at: a.pdf. Muennig P, Cohen AK, Palmer A, Zhu W. The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med. Beyer KM, Layde PM, Hamberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women. Whitley R, McKenzie K. Social capital and review of the literature. Harv Rev De Silva MJ, McKenzie K, Harpham T, Huttly SR. Social capital and mental illness: a systematic review. Epidemiol Community Health. Wyman PA. Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med. Suppl Centers for Disease Control and Prevention. Principles of community engagement. TSDR Committee on Community Engagement. 199?. Branas CC, Cheney RA, MacDonald JM, Tam VW, Jackson TD, Ten Have TR. A difference-in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol. Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald JM. Urban blight remediation as a cost-beneficial solution to firearm violence. Am Public Health. 2016;106l12lz21SB-2164. Ba ndura A. Socialfoundations of thought and action: A social cognitive theory. Prentice-Hall, Inc; 1986. Pollock LR, Williams JM. Problem-solving in suicide attempters. Med. Herman KC, Borden LA, Reinke WM, Webster-Stratton C. The Impact of the Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q. Knox MS, Bu rkhart K, Hunter KE. ACT against violence parents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. 2010. 48 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. Wasserman D, Hoven CW, Wasserman C, et al. School-based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. Lancet. Wilcox HC, Kellam 56, Brown CH, et al. The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend. 2008;95 Suppl 1:560-73. Kellam 56, Brown CH, Pod uska JM, et al. Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend. 2008;95 Suppl 1:55-528. Webster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation ofthe incredible years teacher and child training programs in high-risk schools. Journal of Child and Webster-Stratton CH, Reid MJ, Beauchaine T. Combining parent and child training for young children with ADHD. Journal of Clinical Child 82 Adolescent Reid MJ, Webster-Stratton C, Hammond M. Follow-up ofchildren who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy. Webster-Stratton C, Hammond M. Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin Webster?Stratton C, Reid MJ, Hammond M. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child Spoth RL, Guyll M, Day 5X. Universal family-focused interventions in alcohol?use disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Alcohol. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999?2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics; 2016. Jobes BA. The Collaborative Assessment and Management ofSuicidality (CAMS): an evolving evidence based clinical approach to Suicidal risk. Suicide Life Threat Behav. Wilcox HC, Wyman PA. Suicide Prevention Strategies for improving Population Health. Child Adolesc Clin Isaac M, Elias B, Kat: LY, et al. Gatekeeper training as a preventative intervention for suicide: a systematic review. Can Gould MS, Munfakh JL, Kleinman M, Lake AM. National suicide prevention lifeline: enhancing mental health care for suicidal individuals and other people in crisis. Suicide Life Threat Behav. 35. Gilbody S, Bower P, Fletcher J, Richards D, Sutton AJ. Collaborative care for depression: a cumulative meta?analysis and review of longer?term outcomes. Arch int Med. 14?232 1. Archer i, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews. 2012;10lArt No Bruce ML, Ten Have TR, Reynolds CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA. Inagaki M, Kawashima Kawanishi C, et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Affect Disord. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. Walrath C, Garraza LG, Reid H, Goldston DB, McKeon R. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. 49 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 1149. Gould MS, Kalafat J, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. Unutzer J, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc. Comtois KA, Jobes DA, 5. O'Connor S, et al. Collaborative assessment and management of suicidality feasibility trial for next-day appointment services. Depress Anxiety. Linehan MM, Comtois KA, Murray AM, et al. Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Diamond GS, Wintersteen MB, Brown GK, et al. Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. JArn Acad Chiia?Aa?aiesc 2010;49i2ileZ-131. Ru benstein LV, Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Systems, Heaith. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buii World Heaith Organ. Motto JA, Bostrom AG. A randomized controlled trial of postcrisis suicide prevention. Serv. Hassanian?Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self~poisoning. Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3?Month Intervention. Suicide Life Threat Behav. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects ofsuicide bereavement on mental health and suicide risk. Lancet Etzersdorfer E, Sonneck G. Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res. Niederkrotenthaler T, Sonneck G. Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21 Szumilas M, Kutcher S. Post-suicide intervention programs: A systematic review. Can Pubiic Heaith. Bohanna I, Wang X. Media guidelines for the responsible reporting of suicide: a review of effectiveness. Crisis. Visser VS, Comans TA, Scuffham PA. Evaluation Of The Effectiveness Community-Based Crisis Intervention Program For People Bereaved By Suicide. Jaurnai of Community 28. Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav. 50 174. 175. 176. 177. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. 2010;341:c5841. Centers for Disease Control and Prevention. National Violent Death Reporting Svstem. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2017. Available online: The National Center for the Review Prevention of Child Deaths. U.S. Child Death Review Programs. Centers for Disease Control and Prevention, Brener ND, Karin L, et al. Methodology of the Youth Risk Behavior Surveillance System-?013. MMWR Recomm Rep. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment bene?t programs 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity tows 1/ v? Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps suicide care Healthcare Telemental health (TMH) Social services Safer suncide care through systems change Henry Ford Perfect Depression Care (Pre~ V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt Government (local . . . intervening at hot spots state) Create - Safe storage practices v/ - protective Public Health environments Emergency Department Counseling on Healthcare 52 Organizational policies and culture Best Available Evidence Together for Life 1/ US Air Force Suicide Prevention Program v? Correctionoi suicide prevention 1/ BusinessiLabor Jus?ce Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Government {locaL state) Promote cannectedness Alcohoi outiet density Business/labor Peer norm programs Public Health Sources ofStrength Education Communitv engagement activities Public Health Greening vacant urban spaces Government {local} Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mentoi Heoith Program Education Good Behavior Gome Parenting skill and family relationship approaches The incredibie Years Public Health Fe?luretinn 53 Strengthening For Parents and Youth 10-14 Best Available Evidence Identify and at-risk Gatekeeper training Public Health Appiied Suicide Intervention Skin's Training Healthcare Garret Lee Smith Federai Grant Program Crisis Intervention Public Health support people Nationai Suicide Prevention Lifeiine Social Services Treatment for people at risk of suicide improving Mood Promoting Access to Coiiaborative Treatment Coiioborative Assessment and Management of Suicidoiity (CAMS) Diaiecticai Behaviorai Therapy (DBT) Healthcare Social Services Justice Attachment-Based Famiiy Therapy (ABFT) Transiating initiatives for Depression into E??ective Soiutions project Treatment to prevent re~attempts ED Brief intervention with Foiiow-up Visits Health ca re Active foiiow?up contact approaches EST for Suicide Prevention Social services 54 Best Available Evidence Postvention Lassen harms Stond?y Response Service 1/ 1/ ea care and prevent future risk Safe reporting and message about suicide Public Health 1/ Media Gurdeimes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. Preventing Suicide: A Technical Package of Policy, Programs. and Practices Prepared by: Deb Stone, Set), MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical' Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hoary, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casev Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Christine Schuler National Institute for Occupational Safetv and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellvson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and deliverv of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priorityr to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of 'a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome.1 Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific wavs to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injurv.2 Self-directed violence may be suicidai or non- suicidai in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fatolself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs.Em In 2015 (the most recent year of available death data), suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes.5 In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S. Overall suicide rates increased 28% from 2000 to 2015.5 Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; the fourth leading cause among persons in their 405, and the seventh leading cause among persons in their 505.5 Suicide rates vary by race/ethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.5 Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 45% from 2000 to 2015, with steep increases seen among both males and females aged 45-64 years;5 veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; 3 and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority peersf?rm Suicides reflect only a portion of the problem.11 Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts} than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.11 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation).11 Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature.?12 Risk and protective factors for suicide exist at each level. For example, risk factors include: 0 Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress Community level: inadequate community connectedness, barriers to health care ie.g., lack of access to providers and medications) Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness.M it is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio?cultural and economic status.? Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means.? These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts.?19 Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.? Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide attempts.1512'a The effects ofviolence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide.21'23 Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one's community,?1 school,25 family,26 caring adults,2128 and pro-social peers29 can enhance resilience and help reduce risk for suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities.3??33 In an early study, Crosby and Sacks? estimated that 7% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et al.35 found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/'or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt.?5 Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief,37 stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide.?39 Less discussed but no less important, are the financial and occupational effects for those left behind.? The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alone.?m Adjusting for potential under-reporting of suicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013.41 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.?1 The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable} While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approachfr?? Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, and labor; 3?43 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized 9 controlled trial or quasi-experimental design) evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality ofthe available evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cutting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports . Strengthen household financial security 0 Housing stabilization policies 10 Strengthen access and delivery of suicide 0 Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas I Safer suicide care through systems change Create protective environments 0 Reduce access to lethal means among persons at- risk of suicide 1! Organizational policies and culture 0 Community?based policies to reduce excessive alcohol use Peer norm programs Community engagement activities Social-emotional learning programs Parenting skill and family relationship programs Gatekeeper training Crisis Intervention Treatment for people at-risk of suicide Treatment to prevent re-attempts Postvention Safe reporting and messaging about suicide Promote CDi'l nectedness Teach coping and problem-solving skills Identify and support people at-risk Lessen harms and prevent future risk 0 it is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teoch Coping and Problem- Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced.44 ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to 11 supportive and effective care and provide opportunities to develop individual and community resilience. identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. in practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program materials, training and technical assistance] can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts}. The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 12 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working? age individuals 25 to 64 years Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation ofsuch financial stress may increase an individual's riskfor suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide.50 For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates 13 I Reductions in eviction rates I Reductions in emotional distress - Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal-State Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss.51 An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits mean level: $7,990 per person in 0.5. constant dollars; 5'1 The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment {less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk.52 Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliffsa examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin 8: Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year.53 Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. 1d Housing stabilization policies. The Notional Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularlyr among working-aged adults?"fl Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 {before the housing crisis began] to 2010 (after it had Most of these suicides occurred prior to the actual loss ofthe decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide55v56 and the level of risk conferred by different types of mental illness previous research indicates that mental illness is an important risk factor for suicide?!? State-level suicide rates have also been found to be correlated with general mental health measures such as depression.51152 Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions.53 Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services.?54 Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention.3 Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care.55 Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: 0 Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on parwith coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatient/outpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. If a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others], then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health providers; this shortage is particularly severe among low-income urban and rural communities.? There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and 16 federal programs loan repayment programs} and expanding the reach of health services through telephone, video and web~based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People Alt-Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes I Increases in access to mental health services I Increase in utilization of mental health services I Reductions in of mental illnesses and suicidality I Reductions in rates ofsuicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and service utilization. Using data from this survey, Harris, Carpenter, Baa 5?3 found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang 52 examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of S92 suicides per year.52 Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps (NHSC), which offers 17 financial incentives to attract mental/behavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas (HPSAs) in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps.59 Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Tei'ementoi heoith (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities} to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deaths among other outcomes?0 Further, Mohr and colleagues?1 conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face?to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy. Thus, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered the Perfect Depression Core program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide.n An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 {110.3 to 47.6 per 100,000 population; 18 p<.04) with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 19 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes."5 Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide}:12 For example, rates of suicide are high among middle- aged adults who comprise 42.6% of the workforce; 7'5 among certain occupational groups farming, fishing, forestry, and construction; 35"? and among people in detention facilities jail, prison), to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help?seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not.?F9 Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transitionigtf?5 Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. - Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutesf?r? and 2) people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access?gr? Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 0 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops}, railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing 20 I signs and telephones to encourage individuals who are considering suicide, to seek help?? 0 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms-- ocked in a secure place leg, in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts.30'91 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments leg. residential settings}. Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help}, skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors leg. depression, social Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides.93'94 Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts.95 Potential Outcomes Increases in safe storage of lethal means Reductions in rates of suicide Reductions in suicide attempts Reductions in suicide deaths Increases in help-seeking Reductions in alcohol-related suicide deaths 21 Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hatspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicideg?vg? For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year.? Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to other jumping sites?? Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased five-fold??3 Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 3? counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman, Mueller, Riedy, Dowd, Villaveces, Prodzinski, Nakagawara, Howard, Thiersch, Ha rruff 99 found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one's own.91 Another program, the Emergency Department Coanseiing an Access to l.ethail Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al.3?3' found that at post-test 76% (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-tests? Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and 22 all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking.10D Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 5.4 suicides per 100,000 population per year compared to an 11% increase in the control city 29.0 per 100,000; 100 Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training}, and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community.101 Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide.1?1The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch} found significantly lower rates of suicide after the program was launched than before."8 These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years?8 Finally, while the evidence is still being built for suicide prevention in correctionai facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behaviorfsuicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year.102 Other similar programs have seen declines in suicide both in the United States and internationally.103 Community?based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density and risk factors for 23 suicide, such as interpersonal violence and social connectednessm?i'1m suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that greater density of bars, specifically, is related to greater suicide and suicide attempts, particularly in rural areas.? 24 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality.1?9 Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others.?? Social connections can be formed within and between multiple levels ofthe social ecology,? for instance between individuals peers, neighbors, co-workers}, families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement), and improved mental health. Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole.ml Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. 0 Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- Ievel beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. 0 Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. 25 Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors - Increases in referrals for youth in distressed I Increases help-seeking behaviors 0 Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. I Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools {6 metropolitan, 12 rural), Wyman, Brown, LolVIurray, Schmeelk-Cone, Petrova, Yu, Walsh, Tu, Wang 29 found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders.? I Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2003. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other benefits, including reductions in firearm assaults and vandalism?lil14L 26 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors.3 Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,115 surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness} characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters.115 Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem?solving skills. . Social-emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work?? I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 27 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors li.e., depression, anxiety, conduct problems, substance abuse) - Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse.113 Social?emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems.119 In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to mm were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group relative risk fell by 49.6%.119 Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior?? Two cohorts of youths participated in the program in 1985-86 and 1986-87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the 1336 indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation and 28 suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The (336 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between (586 and the control interventions.120 The researchers believed this may have been due to a lack of implementation fidelity, including less mentoring and monitoring of teachers. 636 was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the first cohort of students. Results for the second cohort were generally smaller but in the desired direction.121 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (in is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence (all protective factors for suicide}.117 The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the program on reducing internalizing such as anxiety and depression, and child conduct problems?;123 The program is also associated with improved problem-solving and conflict management; these skills were maintained at Additionally, the program demonstrated greater benefits in mother-rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.117 Additionally, Strengthening Families 10?14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide?? Strengthening Families has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families?? 29 Identify and Support People At-Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority groups?-T??-m-123 Supporting people at?risk requires proactive case finding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible Internet-based services when appropriate] and engaging people in evidence-based care through such measures as collaborative treatment), remain key challenges. Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of pepple who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access SUpportive services. Approaches The following approaches focus on identifying and supporting people at increased risk ofsuicide. - Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. in Treatment for people at-risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with problem-solving and emotion regulation. Treatment usually takes place in a one-on-one or group format between patients and clinicians and can vary in duration from several weeks to 30 ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/or integrated care linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk.125L131 Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e?mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy andfor group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12913-2594: reattempt within a year, and of attempt survivors die by suicide within 1 to 5 years of their initial attempt.132 Potential Outcomes Reductions in suicide attempts Reductions in suicide rates Reductions in suicide deaths Reductions in of mental illnesses Reductions in suicidal ideation Reductions in mental health-related sequelae Reductions in re-attempts Increases in connectedness Improvements in coping skills Increases in identification of individuals at-risk for suicidal behavior Increases in treatment engagement and compliance by at-risk individuals Increases in community members trained to identify at-risk individuals Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Skiils Training (ASIST) is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, 31 and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman133 evaluated the training across the Notionoi Suicide Prevention Lifeiine network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, the researchers found that callers who spoke with ASiST?trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call, compared to callers who spoke to non-ASiSTtrained counselors. Counselors trained in ASiSTwere also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in did not result in more comprehensive suicide risk assessments than usual care training.133 Gatekeeper training has also been a primary component of the Garret Lee Smith {615) Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GLS trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation.134 This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the (315 program than did similar counties that did not implement GLS activities (4.9 fewer attempts per 1000 youths).135 More than ?9,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionai Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 ofthose completed a follow-up assessment between 1 and 52 clays (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1] to follow-up (time Between time 2 {end of the call) to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.135 32 Treatment for people atnrisk of suicide. The improving Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase) by a depression care manager.? The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow- upm'133 relative to patients who received care as usual. Collaborative Assessment and Management of Suicidaiity (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,139 in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual [intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal id eation, overall distress, and feelings of hopelessness at the 12 month follow-up.140 Other examples include Diaiectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DST were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment {23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined.?11 ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety.142 A randomized controlled trial ofABFTfound that suicidal adolescents assigned to ABFTexperienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than 33 did adolescents receiving enhanced usual care {69.2% vs. 34.6%} and at 24 weeks (82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Solutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 70% of primary care patients.143 also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively.?l3 Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively].144 Another example of treatment to prevent re-attempts involves active foiiow?up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years]. In a meta-analysis conducted by lnagaki et al.,132 interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated.132 Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. 34 in a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto, Bostrom 145 found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide id eation and attem pts.145r14? Finally, Cognitive Behavior Theropyfor Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual.1mg 35 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world. Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide.149 Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion?m151 While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care for the bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions) and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. II Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief.152 I Safe reporting and messaging about suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline), and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion. Potential Outcomes I Reductions in suicidal ideation Reductions in suicide attempts 36 I Reductions in distress I Increases in treatment seeking and engagement - Improvements in reporting following suicide 0 Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service (StondBy), provides clients with face?to-face outreach and telephone support through a professional crisis response team. Site coordinators deveIOp customized case management plans, referring clients to other existing community services matched to their needs.153 In a study by Visser, Comans, Scuffham 153 StondBy clients were significantly less likely to be at high risk for suicidality (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors self-refer for Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendationsfor Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually.?1 Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide.155 Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports.155 37 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Deiivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 38 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 39 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andlor medicolegal standards.2 uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems.2 Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of-death data from death certificates.155 is a state-based surveillance system {currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches?? Data from state and local Child Death Review teams158 and Suicide Death Review Teams {which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department).5 in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth 40 have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose.159 The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies.159 The Notional Survey on Drug Use and Health is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk.150 It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, 41 like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideallv would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantlv, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will plav a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. In closing, and in keeping with a message of resilience as spoken bv those with lived experience, ?hope, help, and healing is possible.? 42 References 10. 11. 12. 13. 14. 15. 16. 18. 19. Frieden TR. Six components necessary for effective public health program implementation. Am Public Health. Crosby AE, Ortega L, Melanson C. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press; 2014. US. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, DC: 2012. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web- Based Injury Statistics Query and Reporting System Available online: KE, Canham-Chervak M, Black SA, Dada EO, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2007*8. injury Prevention. Lineberry TW, O'Connor 55. Suicide in the US Army. Mayo Clinic Proceedings. McIntosh WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholovv B. Suicide Rates by Occupational Group 17 States, 2012. MMWR Morb Mortal Wkly Rap. Kann L, McMa nus T, Harris WA, et al. Youth Risk Behavior Surveillance - United States, 2015. Morbidity and mortality weekly report. Surveillance summaries (Washington, D. C. :2002}. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: Evidence from a national study. American Journal ofpublic health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer .I. Suicidal thoughts and behaviors among adults aged 218 yea rs--United States, 2008-2009. Morbidity and mortality weekly report. Surveillance summaries (Washington, D.C. 2002). Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lazano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56. Bossa rte RM, Karras E, Lu N, et al. Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public health reports (Washington, D. C. Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of a?ective disorders. Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA. 2001;286l24jz3089e3096. Felitti VJ, Anda RF, Nordenberg D, et aI. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. American journal of preventive medicine. 1998;14l4jz245?258. Klomek AB, Sourander A, Gould M. The association of suicide and bullying in childhood to young adulthood: a review of crossusectional and longitudinal research findings. The Canadian Journal of Leeb RT, Lewis T, Zolotor A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. World Health Organization. Global and Regional Estimates ofViolence Against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence. Geneva: World Health Organization; 2013. 43 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. Bellis MA, Hughes K, Leckenby N, et al. Adverse childhood experiences and associations with health- harming behaviours in young adults: surveys in eight eastern European countries. Buli Worid Heaith Organ. Haegerich TM, Dahlberg LL. Violence as a public health risk. American Journai of Lifestyie Medicine. 2011:1559827611409127. Wilkins N, Tsao B, Hertz M, Davis R, Klevens J. Connecting the dots: an overview of the links among multiple forms of violence. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention,- 2014. Hamby S, J. The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociology. New York, NY: Springer; 2013. Kleiman EM, Riskind JH, Schaefer KE, Weingarden H. The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis. Carter M, McGee R, Taylor B, Williams 5. Health outcomes in adolescence: associations with family, friends and school engagement. JAdolesc. Maimon D, Browning CR, Brooks-Gunn J. Collective efficacy, family attachment, and urban adolescent suicide attempts. Heaith Soc Behav. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse. Losel F, Fa rrington DP. Direct protective and buffering protective factors in the development of youth violence. American journai of preventive medicine. 2012;43(2 Suppl 1):58v523. vaan PA, Brown CH, LoMurray M, et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Pubiic Heaith. Dunne EJ, McIntosh JL, Dunne~Maxim K, eds. Suicide and its aftermath: Understanding and counseling the survivors. New York: Norton; 1987'. Mishara BL, ed The Impact of Suicide. New York: Springer; 1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to Grief, Trauma, and Distress After a Suicide: US. National Guidelines. Washington, DC. Author; 2015. Crosby AE, Sacks JJ. Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and Life-Threatening Behavior. Cerel J, Maple M, De Venne A, Moore M, Flaherty C, Brown M. Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports. 2016;131i1l. Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self?harm and suicide attempts. Suicide Life Threat Behav. Mitchell AM, Kim Y, Prigerson HG, Mortimer-Stephens M. Complicated grief in survivors of suicide. Crisis. Sudak H, Maxim K, Carpenter M. Suicide and stigma: a review of the literature and personal reflections. Acad Cerel J, McIntosh JL, Neimeyer RA, Maple M, Marshall D. The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and Life-Threatening Behavior. Florence C, Simon T, Haegerich T, Luo F, Zhou C. Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortai'ityr weeidy report. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr., Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav. 44 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. Silverman MM, Maris RW. The prevention of suicidal behaviors: An overview. Suicide and Life- Threatening Behavior. National Action Alliance for Suicide Prevention. Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. 2014. Turecki G. Epigenetics and suicidal behavior research pathways. American journai of preventive medicine. Edwards RW, Jumper-Thurman P, Plested BA, Oetting ER, Swanson L. Community readiness: Research to practice. Journai of community Hawkins JD, Catalano RF, Kuklinski MR. Communities that care. Encyciopedia of Criminoiogy and Criminai Justice: Springer; 2014:393-408. Plested BA, Edwards RW, Jumper-Thurman P. Community readiness: A handbook for successful change. Fort Coffins, C0: Tri-Ethnic Centerfor Prevention Research. 2006. Luo F, Florence CS, Quispe-Agnoli M, Ouyang L, Crosby AE. Impact of business cycles on US suicide rates, 1928-2007. Am Pubiic Heaith. Fowler KA, Gladden RM, Vagi Ki, Barnes 1, Frazier L. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am i Pahiic Heaith. Stack S, Wasserman Economic strain and suicide risk: a qualitative analysis. Suicide and WEE-threatening behavior. Cylus J, Glymour MM, Avendano M. Do generous unemployment benefit programs reduce suicide rates? A state fixed?effect analysis covering 1968-2008. Am .i Epidemioi. Classen Tl, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heoith Econ. Flavin P, Radcliff B. Public policies and suicide rates in the American states. Socioi indicators Research. Houle JN, Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith. Owens 0. Fatal and non-fatal repetition of self?harm: Systematic review. British ,iournai of Olfson M, Gerhard T, Huang C, S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Arsenault?Lapierre G, Kim C, Turecki G. diagnoses in 3275 suicides: a meta?analysis. BMC 2004;4:37. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A metadanalysis. Br} Tyrer P, Reed GM, Crawford MJ. Classification, assessment, prevalence, and effect of personality disorder. Lancet. Harris EC, Barraclough B. Excess mortality of mental disorder. lBr..f Mark TL, Shern DL, Bagalman JE, Cao Z. Ranking America?s mental health: An analysis of depression across the states. Aiexandria, VA: iVientall Heaith America. 2007. Lang M. The impact of mental health insurance laws on state suicide rates. Heaith Econ. 88. Kessler RC, Demler 0, Frank RG, et al. Prevalence and treatment of mental disorders, 1990 to 2003. Engii Med. Cunningham PJ. Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff Coffey CE. Building a system of perfect depression care in behavioral health. it Comm Qua! Patient Saf. 45 66. 67. 68. 69. 70. 72. 73. 24. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. U.S. Department of Health and Human Services Health Resources and Services Administrations. Designated Health Professional Shortage Areas Statistics, As of November 4, 2016. 2016. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author; 2014. Harris KM, Carpenter C, Bao Y. The effects of state parity laws on the use of mental health care. Med Care. U.S. Department of Health and Human Services Health Resources and Services Administrations. NHSC Clinician Retention: A Story of Dedication and Commitment. 2016. Hailey D, Roine R, Ohinmaa A. The effectiveness of telemental health applications: a review. The Canadian Journal of Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect oftelephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice. Coffey CE. Pursuing perfect depression care. Serv. Coffey CE, Coffey Ml, Ahmedani BK. An update on perfect depression care. Serv. Coffey M, Coffey C, Ah medani BK. Suicide in a health maintenance organization population. JAMA Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Reports. Toosi M. Labor force projections to 2024: The labor force is growing, but slowly. In: Statistics ed. Washington, D.C.: Bureau of Labor Statistics; 2015:1?33. Han B, Crosby AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ldeation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive Knox KL, Pflanz S, Talcott GW, et al. The US Air Force suicide prevention program: implications for public health policy. Am Public Health. National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Blueprint for Workplace Suicide Prevention. Washington, DC. Author; 2015. Runyan CW, Becker A, Brandspigel 5, Barber C, Trudeau A, Novins D. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med. Miller M, Warren M, Hemenway D, Azrael D. Firearms and suicide in US cities. in} Prev. 2015;21iellze116-119. Crosby AE, Espitia~Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Alcohol: Science, Policy and Public Health. 2013:190. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: a gender?stratified analysis of the National Violent Death Reporting System. in} Prev. 2013;19illz38-43. Beautrais AL, Gibb SJ, Fergusson DM, Horwood U, La rkin GL. Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal ldeation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Journal of correctional health care the of?cial journal of the National Commission on Correctional Health Care. Simon OR, Swann AC, Powell KE, Potter LB, Kresnow Ml, D'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32{1 Deisenhammer EA, CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 46 38. 89. 90. 91. 92. 93. 94. 95. 96. 97. 93. 99. 100. 101. 102. 103. 104. 105. 106. 107. 103. 109. 110. Hawton K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. 2007;28lSuppl. ljz4-9. Yip P, Caine E, Yousuf S, Chang S-S, Wu K, Chen Means restriction for suicide prevention. l_ancet. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health. Rowhani-Rahbar A, Simonetti JA, Rivara FP. Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiol Rev. Hayes LM. Suicide prevention in correctional facilities: reflections and next steps. lnternationaljournal of law and Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction. Escobedo LG, Ortiz M. The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Cherpitel CJ, Borges GLG, Wilcox HC. Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research. 2004;28{5 Pirkis J, Too LS, Spittal Ml, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Ami Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ Grossman DC, Mueller BA, Riedy C, et al. Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA. 2005;293l?jz707?714. Mishara BL, Martin N. Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide. Knox KL, Litts DA, Talcott GW, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. Haves LM. Prison Suicide: An Overview and a Guide to Prevention. The Prison Journal. 455. Barker E, Kblves K, De Leo D. Management ofSuicidal and SelfaHarming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol?related damage. I. The distribution of consumption model. JStud Alcohol. Gruenewald Pi, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res. Lipton R, Gruenewald P. The spatial dynamics of violence and alcohol outlets. JStad Alcohol. Lippy C, DeGue S. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Violence, 8: Abuse. Johnson FW, Gruenewald Pi, Remer LG. Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res. 2009;33l12jz2124-2133. Durkheim E. Suicide: A study in sociology Spaulding 8: G. Simpson, trans.). Glencoe, it: Free Press.(Original work published 1897). 1897l1951. Centers for Disease Control and Prevention. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. 2009; 47 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 12 6. 127. 123. 129. 130. Available at: a.pdf. Muennig P, Cohen AK, Palmer A, Zhu W. The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Socialscience 8: medicine (1982). Beyer KM, Layde PM, Ha mberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women. Branas CC, Cheney RA, MacDonald JM, Tam VW, Jackson TD, Ten Have TR. A difference-in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol. Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald JM. Urban blight remediation as a cost-beneficial solution to firearm violence. American journal ofpuhlic health. Ba ndura A. Socialfaundations of thought and action: A social cognitive theory. Prentice-Hall, Inc; 1986. Pollock LR, Williams JM. Problem-solving in suicide attempters. Med. Herman KC, Borden LA, Reinke WM, Webster?Stratton C. The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring lnternalizing Q. Knox MS, Bu rkhart K, Hunter KE. ACT against violence parents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journal of Family issues. Wasserman D, Hoven CW, Wasserman C, et al. School?based suicide prevention programmes: The SEYLE cluster~randomised, controlled trial. The Lancet. Wilcox HC, Kellam 56, Brown CH, et al. The impact of two universal randomized first? and second?grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend. 2008;95 Suppl 1:56.033. Kellam 56, Brown CH, Poduska JM, et al. Effects ofa universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend. 2008;95 Suppl 1:55?528. Wehster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high?risk schools. Journal of child and Webster?Stratton CH, Reid Ml, Beauchaine T. Combining parent and child training for young children with ADHD. Journal of Clinical Child 8: Adolescent 201 Reid MJ, Webster-Stratton C, Hammond M. Followdup ofchildren who received the Incredible Years intervention for oppositionalmdefiant disorder: Maintenance and prediction of 2?year outcome. Behavior Therapy. Webster-Stratton C, Hammond M. Treating children with early?onset conduct problems: a comparison of child and parent training interventions. Consult Clin Webster-Stratton C, Reid MJ, Hammond M. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child Spoth RL, Guyll M, Day SX. Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. Stud Alcohol. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics; 2016. Gilbody S, Bower P, Fletcher J, Richards D, Sutton AJ. Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternal medicine. 2321. Archer l, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews. 2012;10lArt. No. CDOOESZS). 48 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. Bruce ML, Ten Have TR, Reynolds CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. lama. lnagaki M, Kawashima Y, Kawanishi C, et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Journai of affective disorders. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. Walrath C, Garraza LG, Reid H, Goldston DB, McKeon R. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 1149. Gould MS, Kalafat J, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part2: Suicidal callers. Suicide ti?fe Threat Behav. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. Bil/ii. UnutzerJ, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. Am Geriatr Soc. Jobes DA. The Collaborative Assessment and Management of Suicidaiity (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav. Comtois KA, Jobes DA, 5. O'Connor 5, et al. Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depression and anxiety. 972. Linehan MM, Comtois KA, Murray AM, et al. Two-year randomized controlled trial and follow?up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 2006;63i7lz757r766. Diamond GS, Wintersteen MB, Brown GK, et al. Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Acad ChiidAdoiesc Rubenstein LV, Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Systems, Heaith. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buii Worid Heaith Organ. Motto JA, Bostrom AG. A randomized controlled trial of postcrisis suicide prevention. Serv. Hassanian?Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. British .iournaiI of Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander iE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects of suicide bereavement on mental health and suicide risk. Lancet Etzersdorfer E, Sonneck G. Preventing suicide by influencing mass-media reporting: The Viennese experience, 1980?1996. Arch Suicide Res. 49 151. 152. 153. 154. 155. 15 5. 157. 158. 159. 160. Niederkrotenthaler T, Sonneck G. Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust Al 2] Szumilas M, Kutcher S. Post-suicide intervention programs: a systematic review. Con} Public Health. Visser VS, Comans TA, Scuffham PA. Evaluation Of The Effectiveness Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 28. Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behov. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. BMJ. 2010;341:6841. Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. National Violent Death Reporting System. Atlanta, Available online: The National Center for the Review Prevention of Child Deaths. U.S. Child Death Review Programs. Centers for Disease Control and Prevention, Brener ND, Kann L, et al. Methodology of the Youth Risk Behavior Surveillance System?r2013. MMWR Recomm Rep. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016; 50 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment bene?t programs 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity tows 1/ v? Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps suicide care Healthcare Telemental health (TMH) Social services Safer suncide care through systems change Henry Ford Perfect Depression Care (Pre~ V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt Government (local . . . intervening at hot spots state) Create - Safe storage practices v/ - protective Public Health environments Emergency Department Counseling on Healthcare 51 Organizational policies and culture Best Available Evidence Together for Life 1/ US Air Force Suicide Prevention Program v? Correctionoi suicide prevention 1/ BusinessiLabor Jus?ce Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Government {locaL state) Promote cannectedness Alcohoi outiet density Business/labor Peer norm programs Public Health Sources ofStrength Education Communitv engagement activities Public Health Greening vacant urban spaces Government {local} Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mentoi Heoith Program Education Good Behavior Gome Parenting skill and family relationship approaches The incredibie Years Public Health Fe?luretinn 52 Strengthening Families 10?14 Best Available Evidence Identify and at-risk Gatekeeper training support people Public Health Applied Suicide intervention Skills Training Healthcare Garret Lee Smith Federal Grant Program 1/ v? Crisis Intervention Public Health . A . . . Social Services National Prevention Lifeline 1/ Treatment for people at risk of suicide improving Mood Promoting Access to f, Collaborative Treatment Collaborative Assessment and Management Health ca re of5uicidality (CAMS) Dialectical Behavioral Therapy (DBT) v? ervlces Justice Attachment-Based Family Therapy (ABFT) 1/ Translating initiatives for Depression into E??ective Solutions project Treatment to prevent re~attempts ED Brief intervention with Follow-up Visits Health to re Active fallow?up contact approaches v" Social services EST for Suicide Prevention 1/ 53 Best Available Evidence Postvention Lassen harms Stond?y Response Service 1/ 1/ ea care and prevent future risk Safe reporting and message about suicide Public Health 1/ Media Gurdeimes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. Preventing Suicide: A Technical Package of Policy, Programs. and Practices Prepared by: Deb Stone, Set), MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical' Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Hoary, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casev Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Christine Schuler National Institute for Occupational Safetv and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellvson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and deliverv of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessening harms and preventing future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention1 and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome.3 Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific wavs to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention is part of a broader class of behavior called self-directed vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injurv.?El Self-directed violence may be suicidoi or non- suicidai in nature. For the purposes of this document, we refer only to behavior where suicide is intended: Suicide is a death caused by self?directed injurious behavior with any intent to die as a result of the behavior. Suicide attempt is defined as a non?fotoiselfvdirected and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs.1r5 In 2015 [the most recent year of available death dataj, suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes.?5 In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the I:Zlverall suicide rates increased 28% from 2000 to 2015-5 Suicide is a problem throughout the life span; it is the second leading cause of death among those aged [1&0-34 years; the fourth leading cause among persons to 4-1, and?the seventhfifth leading cause among persons Meir?5054564 and eighth leading cause among persons 55-6451 Suicide rates vary by racei'ethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non?Hispanic American lndiaanlaska Native and non? Hispanic White population groups. In 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.E Other population groups disproportionately impacted by suicide include [middle-aged adults (whose rates increased 45% from 2000 to 2015, with steep increases seen among If both males and females (64%1 aged 45-64 years'hlleterans and other military personnel {whose}. suicide rate nearly doubled from 2003 to 2008, surpassing the rate ofsuicide among civilians for the first time in decadeslarg; workers in certain occupational groupsf? and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority Comment IAI: Linda, the field is really used to these age groups. I wouldn't use the 40's and 50's. lthink it gets oonfusing to ask people to make that shift. it ,confused me! Comment We always use 35-64 as middle aged, both in the and in our foa targeting middle aged males. if we go with 35-54, the corresponding percentages would be 33% overall, with increases among and ,females=51.1% i Suicides re?ect only a portion of the problem?? Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide than are fatally injured, and an even greater number are either treated in ambulatory settings le.g., emergency departments) or not treated at all.13 Fo example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adult 5 treated in hospital emergency departments for self-harm injuries, 2? who reported making a suicide attempt, and over 227 who reported seriously considering suicide?r14 Suicide is associated with several risk and Protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time}-5 The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal - is a useful framework for viewing and understanding suicide risk factors identified in the literature.15 Risk and protective factors for suicide exist at each level. For example, risk factors include: 0 Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress Community level: inadequate community connectedness, barriers to health care ie.g., lack of access to providers and medications) Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness}:5 it is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio?cultural and economic status}:5 Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means}:5 These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts.?22 Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.22 Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide attempts?:23 The effects ofviolence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide.?26 Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one's community,? schoolf?S family,29 caring adults,30'31 and pro-social peersa'2 can enhance resilience and help reduce risk for suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities.3335 In an early study, Crosby and Sacks? estimated that 7% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death.? In a more recent study, in one state, Cerel et al33 found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/'or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt.39 Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief,? stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide.?42 Less discussed but no less important, are the financial and occupational effects on those left behind.43 The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alonefl3 Adjusting for potential under-reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self?directed violence to be approximately $93.5 billion in 20133"4 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million?i?l The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable.1-s While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach.?l5 Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public}!5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized 9 controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c} meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or cl] evidence from at least one rigorous RCT or quasi?experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating 3 Comment IAI: Can we add a sentence about as more evidence becomes available we will be updating this document? Trying to appease Dan a Jittle further here. Comment A z This may be the language from other packages as well but seems like a wordisl is missing?not only complementary, but potentially Lsynergistic? 1 Comment We [the SP field} usually refers to it this way, not sure if this is impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help?seeking, stigma reduction, depression, connectedness). In terms of the strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher?level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and conducting more evaluations may be necessary to address different population groups. The evidence based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protectiva factors forfuicid? Context and Cross-Cutting Themes strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more belectgl and indicated ]populations persons who have already made a suicide attempt}. kstandard language or can be changed. Preventing Suicide Approach Strengthen household financial security Housing stabilization policies Strategy Strengthen economic supports ll] Strengthen access and delivery of suicide 0 Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas I Safer suicide care through systems change Create protective environments 0 Reduce access to lethal means among persons at- risk of suicide 1! Organizational policies and culture 0 Community?based policies to reduce excessive alcohol use Peer norm programs Community engagement activities Social-emotional learning programs Parenting skill and family relationship programs Gatekeeper training Crisis Intervention Treatment for people at-risk of suicide Treatment to prevent re-attempts Postvention Safe reporting and messaging about suicide Promote CDi'l nectedness Teach coping and problem-solving skills Identify and support people at-risk Lessen harms and prevent future risk 0 it is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teoch Coping and Problem- Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced.46 ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to 11 supportive and effective care and provide opportunities to develop individual and communitv resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identi?ed in this package will be stronglv dependent on how well programs are implemented, as well as the partners and communities in which theyr are implemented. Practitioners in the field mav be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this These planning processes engage and guide communitv stakeholders through a prevention planning process designed to address a communitv?s profile of risk and protective factors with evidencebased programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated bv the data. The readiness of the program for broad dissemination and implementation availabilitv of program materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed eparatel? This package includes strategies where and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing communitv and societal level risks], but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policv or program le.g., workplace policies; treatment to prevent re-attempts). The role of various sectors in the implementation of a strategv or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. Comment IAI: Linda, the Action Alliance 12 is creating an implementation paper that will come out shortly after our technical package. Is there any wav to add that here? Something like forthcoming document on implementation of comprehensive communitv-based suicide prevention for states and communities is forthcoming." Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working? age individuals 25 to 64 years old.?51 Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation ofsuch financial stress may increase an individual's riskfor suicide or may indirectly increase risk by exacerbating related physical and mental health problems.52 Buffering these risks can, therefore, potentially protect against suicide. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. I Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates 13 I Reductions in eviction rates Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household financial Security. The Federal-State Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss.53 An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: 51qu per person in 0.5. constant dollars]. The effects of unemployment benefit programs were also consistent by sex and age group.53 Another US. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 25 weeks}, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk.? Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security leg, transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Ftadcliff55 examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance [Temporary Assistance to Needy Families - and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors ie.g., residential mobility, divorce rate. unemployment rate} at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved. Flavin and Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied?5 At the national level. they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year?5 [Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and dgi?ncreased risk for suicide. JMore evaluation studies are needed to further understand the outcomes impacted by programs such as theseComment IAI: Kristin can you con?rm this. it seems obvious that it's decreasing not Increasing but maybe l?m missing som ething?? Housing stabilization policies. The Cnmmem Kristin, i think this just to help neighborhoods suffering from high rates offoreclosure and abandonment by slowing the ro ram.T in in to ad is re: deterioration of the neighborhoods and providing affordable housing options for low, moderate, E?gs: e.info rams and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslv evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.5. states, so did the state suicide rate, particuiarlv among working?aged adults.SE Another studv of data from 16 U5. states participating in the Notionoi Vioient Death Reporting System found that suicides precipitated bv home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began] to 2010 {after it had peaked].51 Most ofthese suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financiai. foreclosure, and move-out planning and counseling services may help to prevent suicide. 15 Strengthen Access and Delivery ofSuicide Care Rationale While most people with mental health problems do not attempt or die by sulcide?fja and the level of risk conferred by different types of mental illness varies,?61 previous research indicates that mental illness is an important risk factor for suicide?I52 State-level suicide rates have also been found to be correlated with general mental health measures such as depressionml??1 Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions?'E? Lack of access to mental health care is one ofthe contributing factors related to the underuse of mental health servicesF" Identifying ways to improve access to timely, affordable. and quality mental health and suicide care for people in need is a critical component to prevention.5 Additionally. research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care.? Apart from treatment benefits, these approaches can also normalize help?seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns lie, mental health paritylifg?] Benefits and,1 Services covered include such things as the number of visits, co-pays, deductibles, inpatient/outpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. if a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others}, then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary In the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. I Reduce provider shortages in underserved areas. Access to effective and statesof-the-art mental health care is largely dependent upon the training and the size ofthe mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health providers; this shortage is particularly severe among low?income urban and rural communities.? There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and 15 Comment IAI: We should add .3 ref for MH parity law. Kristin can you find? federal programs loan repayment programs} and expanding the reach of health services through telephone, video and web-based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People Alt-Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes I Increases in access to mental health services I Increase in utilization of mental health services I Reductions in of mental illnesses and suicidality I Reductions in rates ofsuicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey on Drug Use and Health (NSDUH) is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and service utilization?? Using data from this survey, Harris, Carpenter, and Baa? found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang 5? examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year.? Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps (NHSC), which offers 17 financial incentives to attract mental/behavioral health clinicians to underserved areas?2 Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas (HPSAs) in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to the NHSC.73 Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Teiementoi health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance.M TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities) to treat a wide range of mental health conditions. it can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide?4 Further, Mohr and colleagues? conducted a meta?analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone- administered compared to patients receiving face-to-face therapy.75 Th us, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered Perfect Depression Core,TE the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide.? An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 8296.75!? Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 (110.3 to 47.6 per 100,000 population; p<.04] with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members 18 who accessed onlyr general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 19 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes."9 Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide}:15 For example, rates of suicide are high among middle- aged adults who comprise 42.6% ofthe workforceso; among certain occupational groupsmval; and among people in detention facilities jail, prison)? to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition.35'9? Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. - Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 35% of people who use a firearm in a suicide attempt will die from the injury}.91 Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutesFZrE and 2) people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access.M95 Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: intervening at Suicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help.96 20 Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms--locked in a secure place le.g., in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior 0 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential settings}. Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help}, skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors leg. depression, social I Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides.99?1?1 Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age.ml These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts.?32 Potential Outcomes Increases in safe storage of lethal means I Reductions in rates of suicide I Reductions in suicide attempts I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths 21 Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the US. and abroad, found associated reduced rates of suicides??!1G3 For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year.MM Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to other jumping sites.1m Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased five-fold?gr105 Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 3? counties in Washington, Oregon, and Missouri, and from 5 trauma centers, researchers found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescentsm? Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own.? Another program, the Emergency Department Counseling on Access to Lethal Means CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. in a pre-post quality improvement project, Runyan et al35 found that at post-test 76% (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le. all (100%) reported guns were currently locked up at post-test.35 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness 22 of existing resources. The program also included an education campaign to improve awareness and help-seeking.1m Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city 29.0 per 100,000;107 Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community.133 Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide.1?3The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years postulaunch) found significantly lower rates of suicide after the program was launched than before.83 These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years.83 Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behavior/suicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide??3 When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following yeah-??9 Other similar programs have seen declines in suicide both in the United States and Community?based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density and risk factors for suicide, such as interpersonal violence and social connectedness,111'114 suggest that measures to 23 reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that greater density of bars, specifically, is related to greater suicide and suicide attempts, particularly in rural areas.115 24 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality.116 Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others?? Social connections can be formed within and between multiple levels ofthe social ecology,15 for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations?am Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement), and improved mental Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a wholem' Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. 0 Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings.122 0 Community engagement activities. Community engagement is an aspect of social capital}23 Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. 25 Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors - Increases in referrals for youth in distressed I Increases help-seeking behaviors 0 Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. I Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools {6 metropolitan, 12 rural), researchers found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement.32 Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult sopport for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders.32 I Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other cross-cutting benefits, including reductions in firearm assaults and 26 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors.122 Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,126 surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness} characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters.127 Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem?solving skills. . Social-emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences substance use} associated with suicide.122 These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work.1253 I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities?-8 Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 27 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors li.e., depression, anxiety, conduct problems, substance abuse) - Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse.129 Social?emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems.130 In a cluster-randomized controlled trial conducted across 10 European Union countries and 163 schools, students in schools randomized to mm were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group, relative risk fell by 49.6%.130 Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior.131 Two cohorts of youths participated in the program in 1985-86 and 1986-87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the 1336 indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation and 28 suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included.?1 The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between (586 and the control interventions.131 The researchers believed this may have been due to a lack of implementation fidelity, including less mentoring and monitoring of teachers. 636 was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the first cohort of students. Results for the second cohort were generally smaller but in the desired direction.132 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (in is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence (all protective factors for suicide}.128 The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the program on reducing internalizing such as anxiety and depression, and child conduct problems.133v13? The program is also associated with improved problem-solving and conflict management; these skills were maintained at Additionally, the program demonstrated greater benefits in mother-rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.123 Additionally, Strengthening For Parents and Youth 10-14 is a program that involves sessions for parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide.133 Strengthening Families has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families.133 29 Identify and Support People At-Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority Supporting people at?risk requires proactive case finding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible Internet-based services when appropriate] and engaging people in evidence-based care through such measures as collaborative treatment}, remain key Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services.65 Approaches The following approaches focus on identifying and SUpporting people at increased risk of suicide. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support peopie at risk.?12 a Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. 1?3 Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Treatment for people at-risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with problem-solving and emotion regulation. Treatment usually takes place in a one-on-one or 30 group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/or integrated care linkage between primary care and behavioral health care) can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide riskm'm Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 12%~25% reattempt within a year, and of attempt survivors die by suicide within 1 to 5 years of their initial attempt?? Potential Outcomes Reductions in suicide attempts Reductions in suicide rates Reductions in suicide deaths Reductions in of mental illnesses I Reductions in suicidal ideation I Reductions in mental health-related sequelae I Reductions in re-attempts I Increases in connectedness I Improvements in coping skills I Increases in identification of individuals at-risk for suicidal behavior I Increases in treatment engagement and compliance by at-risk individuals I Increases in community members trained to identify at-risk individuals I Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Training (ASIST) is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to 31 identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman143 evaluated the training across the Notionoi Suicide Prevention Lifeline network of hotlines over the period 2008-2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, the researchers found that callers who spoke with ASIST?trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call, compared to callers who spoke to counselors. Counselors trained in also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASLST did not result in more comprehensive suicide risk assessments than usual care training.148 Gatekeeper training has also been a primary component of the Garret Lee Smith Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation.149 This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities {4.9 fewer attempts per 1000 youths).15? More than 79,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 ofthose completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1) to follow-up {time Between time 2 (end of the call] to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.151 32 Treatment for people atnrisk of suicide. The improving Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase) by a depression care manager.152 The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow- up152'153 relative to patients who received care as usual. Collaborative Assessment and Management of Suicidaiity (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,140 in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual [intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal id eation, overall distress, and feelings of hopelessness at the 12 month Other examples include Diaiectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DST were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment {23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined.155 ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety.156 A randomized controlled trial ofABFTfound that suicidal adolescents assigned to ABFTexperienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than 33 did adolescents receiving enhanced usual care {69.2% vs. 34.6%} and at 24 weeks (82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Solutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 70% of primary care patients?? also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively?? Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus Another example of treatment to prevent re-attempts involves active foiiow?up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation.?17 These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years]. In a meta-analysis conducted by lnagaki et all? interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated?? Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. 34 in a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto, Bostrom 159 found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide id eation and attempts.150r161 Finally, Cognitive Behavior Theropyfor Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual.162 35 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world.5 Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide.153 Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion.154?155 While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care for the bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions) and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated griefw?5 I Safe reporting and messaging about suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline), and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion?? Potential Outcomes I Reductions in suicidal ideation Reductions in suicide attempts 36 I Reductions in distress I Increases in treatment seeking and engagement - Improvements in reporting following suicide 0 Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service (StondBy), provides clients with face?to-face outreach and telephone support through a professional crisis response team. Site coordinators deveIOp customized case management plans, referring clients to other existing community services matched to their needs.?58 In a study by Visser, Comans, Scuffham 1?53 StondBy clients were significantly less likely to be at high risk for suicidality (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors self-refer for Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendationsfor Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually.?as Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide.1m Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports.1m 37 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the NotionolStrotegyfor Suicide Prevention,1 the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact ofsuicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Cure. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Sicilis to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 38 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 39 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andlor medicolegal standards.?1 uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems.? Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of-death data from death certificates? is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches?:L Data from state and local Child Death Review teamsm and Suicide Death Review Teams {which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department).5 in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth 40 have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose.?3 The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies.?3 The Notional Survey on Drug Use and Health is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk.m It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, 41 like many public health problems, suicide is preventable}5 and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention,35r35r59?34 the release of the first world report on suicide? and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 42 References 10. 11. 12. 13. 14. 15. 16. 18. 19. U.S. Office of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: 2012. National Action Alliance for Suicide Prevention. Action Alliance Priorities. 2017,- Frieden TR. Six components necessary for effective public health program implementation. Am Public Health. Crosby AE, Ortega L, Melanson C. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press; 2014. Web-Based Injury Statistics Query and Reporting System National Center for injury Prevention and Control. Available at: Accessed January 3, 2017. Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. KE, Canham-Chervak M, Black SA, Dada E0, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2007?8. Prev. Lineberry TW, O'Connor SS. Suicide in the US Army. Mayo Clinic Proceedings. McIntosh WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholow B. Suicide Rates by Occupational Group 17 States, 2012. MMWR Morb Mortal Wkly Rep. Kann L, McMa nus T, Harris WA, et al. Youth Risk Behavior Surveillance - United States, 2015. Morbidity and mortality weekly report. Surveillance summaries (Washington, D. C. :2002}. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: Evidence from a national study. Am Public Health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer J. Suicidal thoughts and behaviors among adults aged >l=18 years?~U nited States, 2008-2009. Morbidity and mortality weekly report. Surveillance summaries (Washington, 01:. :2002). Lipari R, Piscopo K, Kroutil LA, Kilrner Miller G. Suicidal thoughts and behavior among adults: Results from the 2014 National Survey on Drug Use and Health. NSDUH Data Review 2015; Dahlberg LL, Krug EG. Violence~a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1?56. Bossarte RM, Ka rras E, Lu N, et al. Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public health reports (Washington, DC. 1924). Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood. Journal of affective disorders. 2004;82l2jz217-225. Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. Jamo. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med. 43 20. 21. 22. 23. 24. 25. 26. 27. 23. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. Klomek AB, Sourander A, Gould M. The association of suicide and bullving in childhood to voung adulthood: a review of cross-sectional and longitudinal research findings. The Canadian Journal of Leeb RT, Lewis T, Zolotor A1. A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. World Health Organization. Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of Intimate Partner Violence and Non-Partner Sexual Violence. Geneva: World Health Organization; 2013. Bellis MA, Hughes K, Leckenbv N, et al. Adverse childhood experiences and associations with health- harming behaviours in young adults: surveys in eight eastern European countries. Bali Worid Heaith Organ. Haegerich TM, Dahlberg LL. Violence as a public health risk. American Journal of Lifestyie Medicine. Wilkins N, Tsao B, Hertz M, Davis R, Klevens J. In: violence Ctdaootlamfo, ed. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2014. Hambv S, I. The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Socioiogy. New York, NY: Springer; 2013. Kleiman EM, Riskind IH, Schaefer KE, Weingarden H. The moderating role of social support on the relationship between impulsivitv and suicide risk. Crisis. Carter M, McGee R, Taylor R, Williams 5. Health outcomes in adolescence: associations with family, friends and school engagement. JAdoiesc. Maimon D, Browning CR, Brooks?Gunn J. Collective efficacy, familv attachment, and urban adolescent suicide attempts. Heaith Soc Behav. Capaldi DM, Knoble NB, Shortt JW, Kim HK. A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse. Losel F, Fa rrington DP. Direct protective and buffering protective factors in the development of youth violence. Am Prev Med. 2012;43i2 Suppl vaan PA, Brown CH, LoMurran,I M, et al. An outcome evaluation of the Sources of Strength suicide prevention program deiivered bv adolescent peer leaders in high schools. Am Puhiic Heaith. 2010;100i9lzl653?1561. Dunne EJ, McIntosh JL, Dunne-Maxim K, eds. Suicide and its aftermath: Understanding and counseiing the survivors. New York: Norton; 198?. Mishara BL, ed The impact ofSuicide. New York: Springer; 1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The Way Forward: Pathways to hope, recovery, and weiiness with insights from iived experience. Washington, D.C. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, D.C. :Author; 2015. Crosby AE, Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav. Cerel 1, Maple M, De Venne A, Moore M, Flahertv C, Brown M. Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic heaith reports (Washington, DC. 1974). 2016;131i1). Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav. Mitchell AM, Kim Y, Prigerson HG, Mortimer-Stephens M. Complicated grief in survivors of suicide. Crisis. Sudak H, Maxim K, Carpenter M. Suicide and stigma: a review of the literature and personal reflections. Acad 44 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 53. 59. 60. 61. 62. 63. 64. 65. Cerel i, McIntosh JL, Neimeyer RA, Maple M, Marshall D. The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide Life Threat Behav. Florence C, Simon T, Haegerich T, Luo F, Zhou C. Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr., Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav. Silverman MM, Maris RW. The prevention of suicidal behaviors: An overview. Suicide Life Threat Behav. Turecki G. Epigenetics and suicidal behavior research pathways. Amif Prev Med. Edwards RW, Jumper-Thurman P, Plested BA, Oetting ER, Swanson L. Community readiness: Research to practice. Journal ofcommunity Hawkins JD, Catalano RF, Kuklinski MR. Communities that care. of Criminology and Criminal Justice: Springer; 2014:393-408. Plested BA, Edwards RW, Jumper-Thurman P. Community readiness: A handbook for successful change. Fort Collins, CO: Tri-Ethnic Center for Prevention Research. 2006. Luo F, Florence CS, Cluispe-Agnoli M, Ouyang L, Crosby AE. Impact of business cycles on US suicide rates, 1928-2007. Am Public Health. Fowler KA, Gladden RM, ?v'agi Kl, Barnes J, Frazier L. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 20052010. Am Public Health. Stack S, Wasserman I. Economic strain and suicide risk: a qualitative analysis. Suicide and life~threatening behavior. Cylus J, Glymour MM, Avendano M. Do generous unemployment benefit programs reduce suicide rates? A state fixed~effect analysis covering 1968?2008. Am Epidemiol. Classen TJ, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass~layaffs and unemployment duration. Health Econ. Flavin P, Radcliff B. Public policies and suicide rates in the American states. Social indicators Research. Houle JN, Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Public Health. Owens D. Fatal and non-fatal repetition of self-harm: Systematic review. British journal Olfson M, Gerhard T, Huang C, S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Arsenault-Lapierre G, Kim C, Turecki G. diagnoses in 3275 suicides: a meta?analysis. BMC 2004,1137. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-a nalysis. Bri Tyrer P, Reed GM, Crawford MJ. Classification, assessment, prevalence, and effect of personality disorder. Lancet. Harris EC, Barraclough B. Excess mortality of mental disorder. BFJ Mark TL, Shern DL, Bagalman JE, Cao Z. Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 200?. Lang M. The impact of mental health insurance laws on state suicide rates. Health Econ. 88. Wang PS, Demler O, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Am Public Health. 45 66. 67. 68. 69. 7D. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. Cunningham PJ. Bevond parity: primary care phvsicians? perspectives on access to mental health care. Heaith Aff (Miilwaod). Coffey CE. Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf. U.S. Department of Health and Human Services Health Resources and Services Administrations. Designated Health Professional Shortage Areas Statistics, As of November 4, 2016. 2016. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the ciinicai workforce: Guidelines for training. Washington, D.C.: Author; 2014. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016; description.htm . Harris KM, Carpenter C, Bao Y. The effects of state paritv laws on the use of mental health care. Med Care. 2006;44i6lz499-505. Health Resources 8; Services Administration. National Health Service Corps. 2017; U.S. Department of Health and Human Services Health Resources and Services Administrations. NHSC Clinician Retention: A Story of Dedication and Commitment. 2016. Hailev D, Roine R, Ohinmaa A. The effectiveness of telemental health applications: a review. The Canadian Journai of Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect of telephone~administered on of depression and attrition: A metava nalvsis. Ciinicai Science and Practice. Coffev CE. Pursuing perfect depression care. Serv. Coffev CE, Coffev MJ, Ahmedani BK. An update on perfect depression care. Serv. Coffev M, Coffev C, Ah medani BK. Suicide in a health maintenance organization population. JAMA Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports. Toosi M. Labor force projections to 2024: The labor force is growing, but slowlv. In: Statistics ed. Washington, D.C.: Bureau of Labor Statistics; 2015183. Han B, Crosbv AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64vears in the United States. Comprehensive Noonan ME. Mortality in State Prisons, 20012014 Bureau afiustice Statistica.f Tabies. 2016;250150lDecember). Knox KL, Pflanz S, Talcott (SW, at al. The US Air Force suicide prevention program: implications for public health policv. Am Pubiic Heaith. National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Biueprintfor Workplace Suicide Prevention. Washington, DC. Author; 2015. Ru nvan CW, Becker A, Brandspigel 5, Barber C, Trudeau A, Novins 0. Lethal Means Counseling for Parents of Youth Seeking Emergencv Care for Suicidality. WestJr Emergr Med. Miller M, Warren M, Hemenwav D, Azrael D. Firearms and suicide in US cities. in} Prev. Crosby AE, Espitia-Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Aicohoi: Science, Poiicy and Public Heaith. 2013:190. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: a gender-stratified analvsis of the National Violent Death Reporting Svstem. in} Prev. 46 89. 90. 91. 92. 93. 94. 95. 96. 97. 93. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. Beautrais AL, Gibb SJ, Fergusson DM, Horwood U, La rkin GL. Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 2009;43. Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal ldeation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Journal of correctional health care the of?cial journal of the National Commission on Correctional Health Care. Elnour AA, Harrison J. Lethality of suicide methods. in} Prev. Simon OR, Swann AC, Powell KE, Potter LB, Kresnow MJ, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001;32l1 Deisenhammer EA, CM, Strauss R, Kemmler G, Hinterhuber H, Weiss EM. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin Hawton K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. 2007;28. Yip P, Caine E, Yousuf 5, Chang S-S, Wu K, Chen Means restriction for suicide prevention. Lancet. 2012;37?9. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic review. BiviC Public Health. Rowhani-Rahbar A, Simonetti JA, Rivara FP. Effectiveness of interventions to Promote Safe Firearm Storage. Epidemioi Rev. Hayes LM. Suicide prevention in correctional facilities: reflections and next steps. internationaljournal of law and Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off?premise and on-premise alcohol outlet density. Addiction. Escobedo LG, Ortiz M. The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Xuan Z, Naimi TS, Kaplan MS, et al. Alcohol Policies and Suicide: A Review of the Literature. Alcohol Clin Exp Res. Cherpitel CJ, Borges GLG, Wilcox HC. Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research. 2004;28l5 Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 2001;35. Grossman DC, Mueller BA, Riedy C, et al. Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA. Mishara BL, Martin N. Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention Suicide. Knox KL, Litts DA, Talcott GW, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ. Hayes LM. Prison Suicide: An Overview and a Guide to Prevention. The Prison Journal. 456. Barker E, Kolves K, De Leo D. Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research. 20141881221240. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol-related damage. I. The distribution of consumption model. JStud Alcohol. 47 112. 113. 114. 115. 115. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. Gruenevvald PJ, Remer L. Changes in outlet densities affect violence rates. Aicohoi Clin Exp Res. Lipton R, Gruenewald P. The spatial dynamics of violence and alcohol outlets. JStudAicohoi. Lippy C, DeGue S. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Vioience, Abuse. 2014:1524838014557291. Johnson FW, Gruenewald Pi, Remer LG. Suicide and alcohol: do outlets play a role? Aicohoi Clin Exp Res. Durkheim E. Suicide: A study in sociology Spaulding 8: G. Simpson, trans). Giencoe, it: Free Press.(Originai work published 1897). 1897/1951. Centers for Disease Control and Prevention. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. 2009; Available at: Strategic Direction Full version- Muennig P, Cohen AK, Palmer A, Zhu W. The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Sociaiscience medicine (1982}. Beyer KM, Layde PM, Ha mberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other femicides? Vioience Against Women. Whitley R, McKenzie K. Social capital and review of the literature. Harvard review of De Silva MJ, McKenzie K, Harpham T, Huttly SR. Social capital and mental illness: a systematic review. Journai of epidemioiagy and community health. 2005 Wyman PA. Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med. 2014;47(3 Suppl 2}15251?256. Centers for Disease Control and Prevention. Principles of community engagement. TSDR Committee on Community Engagement. 1997. Branas CC, Cheney RA, MacDonald JM, Tam VW, Jackson TD, Ten Have TR. A difference?in?differences analysis of health, safety, and greening vacant urban space. Am Epidemioi. Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald JM. Urban blight remediation as a cost-beneficial solution to firearm violence. Am Public Heaith. Ba ndura A. Social foundations of thought and action: A social cognitive theory. Prentice?Hall, Inc; 1986. Pollock LR, Williams JM. Problem-solving in suicide attempters. Med. Herman KC, Borden LA, Reinke WM, Webster?Stratton C. The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring Internalizing Q. Knox MS, Burkhart K, Hunter KE. ACT against violence parents raising safe kids program: Effects on maltreatment-related parenting behaviors and beliefs. Journai of Family issues. 2010. Wasserman D, Hoven CW, Wasserman C, et al. School-based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet. Wilcox HC, Kellam SG, Brown CH, et al. The impact of two universal randomized first- and second-grade classroom interventions on young adult suicide ideation and attempts. Drug Aicohol Depend. 2008;95 Suppl 1:560-73. Kellam 56, Brown CH, Pod uska JM, et al. Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohoi Depend. 2008;95 Suppl 48 133. 134. 135. 135. 137. 133. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. Webster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schools. Journal of child and Webster-Stratton CH, Reid Ml, Beauchaine T. Combining parent and child training for young children with ADHD. Journal of Clinical Child 3: Adolescent Reid Ml, Webster-Stratton C, Hammond M. Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy. Webster-Stratton C, Hammond M. Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin Webster-Stratton C, Reid MJ, Hammond M. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child Spoth RL, Guyll M, Day 5X. Universal family-focused interventions in alcohol-use disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. Stud Alcohol. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics; 2016. lobes DA. The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav. Wilcox HC, Wyman PA. Suicide Prevention Strategies for Improving Population Health. Child and adolescent clinics of North America. 2016. Isaac M, Elias B, Katz LY, et al. Gatekeeper training as a preventative intervention for suicide: a systematic review. Canadian journal of Revue canadienne de 268. Gould MS, Munfakh JL, Kleinman M, Lake AM. National suicide prevention lifeline: enhancing mental health care for suicidal individuals and other people in crisis. Suicide life Threat Behav. 35. Gilbody S, Bower P, Fletcher l, Richards 0, Sutton AJ. Collaborative ca re for depression: a cumulative meta-analysis and review of longersterm outcomes. Archives ofinternal medicine. 2006;166l21lzz3144 2321. Archer J, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. The Cochrane Library. 2012. Bruce ML, Ten Have TR, Reynolds Ill CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. Jama. Inagaki M, Kawashima Kawanishi C, et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Journal of affective disorders. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. Walrath C, Garraza LG, Reid H, Goldston DB, McKeon R. Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 1149. Gould MS, Kalafatl, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ. 49 153. 154. 155. 156. 157. 153. 159. 160. 161. 162. 163. 164. 165. 166. 167. 163. 169. 170. 171. 172. UnutzerJ, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. Am Geriatr Soc. Comtois KA, Jobes DA, 5. O'Connor S, et al. Collaborative assessment and management of suicidality feasibility trial for next-day appointment services. Depression and anxiety. 972. Linehan MM, Comtois KA, Murray AM, et al. Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Diamond GS, Wintersteen MB, Brown GK, et al. Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad ChiidAdoiesc Ru benstein LV, Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Families, Systems, 84 Heaith. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Buii' Worid Health Organ. Motto JA, Bostrom AG. A randomized controlled trial of postcrisis suicide prevention. Serv. Hassanian?Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self~poisoning. British Journai of Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3?Month Intervention. Suicide Life Threat Behov. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects of suicide bereavement on mental health and suicide risk. Lancet Etzersdorfer E, Sonneck G. Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res. 1998;4. Niederkrotenthaler T, Sonneck G. Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 2007;?. Szumilas M, Kutcher S. Post?suicide intervention programs: a systematic review. Can Puhiic Heaith. Bohanna I, Wang X. Media guidelines for the responsible reporting of suicide: a review of effectiveness. Crisis. Visser VS, Comans TA, Scuffham PA. Evaluation Of The Effectiveness Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journai of Community 28. Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. 2010;341:c5841. National ?v'iolent Death Reporting System. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2017. Available at: rs.html. The National Center for the Review 8-: Prevention of Child Deaths. U.S. Child Death Review Programs. 50 173. Centers for Disease Control and Prevention, Brener ND, Kann L, et al. Methodology of the Youth Risk Behavior Surveillance System--2013. MMWR Recomm Rep. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment bene?t programs 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity tows 1/ v? Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps suicide care Healthcare Telemental health (TMH) Social services Safer suncide care through systems change Henry Ford Perfect Depression Care (Pre~ V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt Government (local . . . intervening at hot spots state) Create - Safe storage practices v/ - protective Public Health environments Emergency Department Counseling on Healthcare 52 Organizational policies and culture Best Available Evidence Together for Life 1/ US Air Force Suicide Prevention Program v? Correctionoi suicide prevention 1/ BusinessiLabor Jus?ce Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Government {locaL state) Promote cannectedness Alcohoi outiet density Business/labor Peer norm programs Public Health Sources ofStrength Education Communitv engagement activities Public Health Greening vacant urban spaces Government {local} Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mentoi Heoith Program Education Good Behavior Gome Parenting skill and family relationship approaches The incredibie Years Public Health Fe?luretinn 53 Strengthening For Parents and Youth 10-14 Best Available Evidence Identify and at-risk Gatekeeper training Public Health Appiied Suicide Intervention Skin's Training Healthcare Garret Lee Smith Federai Grant Program Crisis Intervention Public Health support people Nationai Suicide Prevention Lifeiine Social Services Treatment for people at risk of suicide improving Mood Promoting Access to Coiiaborative Treatment Coiioborative Assessment and Management of Suicidoiity (CAMS) Diaiecticai Behaviorai Therapy (DBT) Healthcare Social Services Justice Attachment-Based Famiiy Therapy (ABFT) Transiating initiatives for Depression into E??ective Soiutions project Treatment to prevent re~attempts ED Brief intervention with Foiiow-up Visits Health ca re Active foiiow?up contact approaches EST for Suicide Prevention Social services 54 Best Available Evidence Postvention Lassen harms Stond?y Response Service 1/ 1/ ea care and prevent future risk Safe reporting and message about suicide Public Health 1/ Media Gurdeimes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. Preventing Suicide: A Technical Package of Policy, Programs, and Practices National Center for Injuryr Prevention and Control Division of Violence Prevention Preventing Suicide: A Technical Package of Policy, Programs, and Practices Developed by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, 2017 Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention Atlanta, Georgia Centers for Disease Control and Prevention Anne Schuchat, MD (RADM, USPHS), Acting Director National Center for injury' Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing SuiciderA Technicai Package afPoiicies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Preventing Suicide: A Technical Package of Policy, Programs, and Practices Contents Acknowledgements 5 External Reviewers 5 Overview 7 Strengthen Economic Supports for Families 1 5 Strengthen Access and Delivery of Suicide Care 19 Create Protective Environments 23 Promote Connectedness 27 Teach COping and Problem-Solving Skills 31 Identify and Support People At?Risk 35 Lessen Harms and Prevent Future Risk 41 Sector Involvement 43 Monitoring and Evaluation 45 Conclusion 47 References49 Appendix: Summary of Strategies and Approaches to Prevent Suicide 58 Preventing Suicide: ll Technical Package of Policy, Programs, and Practices 3? Ei' BOD LL \[He Acknowledgements We would like to thank the following individuals who contributed in speci?c ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but de?nitely not least, we extend our thanks and gratitude to all the external reviewers fortheir helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Of?ce for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindman Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Dan Reidenberg Suicide Awareness Voices for Education (SAVE) Christine Schuier National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Preventing Suicide: A Technical Package of Pn?cy, Programs. and Practices Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessening harms and preventing futurg .The strategies represented in this package include those with a focus on preventing suicide from happenin the ?rst place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy forSuicia'e Prevention1 and the National Action Alliance for Suicide Prevention's priority to strengthen community?based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education,justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a speci?c risk factor or outcomeFTechnical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The ?rst component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the speci?c ways to advance the strategy. This can be accomplished through programs, policies, and practices. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision?making in communities and states. Preventing Suicide is a Priority Suicide, as de?ned by the Centers for Disease Control and Prevention (CDC), is part ofa broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury? Self-directed violence may be suicidal or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. . Suicide attempt is de?ned as a non?fatal selfndirected and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs.? In 2015 (the most recent year of available death data}, suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes?" In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes ofdeath sinc '25 in the U.S.7 Overall suicide rates increased 28% from 2000 to 2015.5 Suicide is a problem throughout the life i; it is the third leading cause of death for youth 10?1 4 years of age, the second leading cause of death among people 15r24 and 25-34 years of age; the fourth leading cause among people 35 to 44 years of age, the ?fth leading cause among people ages 4564 and eighth leading cause among people 55-64 years of age.r5 Suicide rates vary by racefethnicity, age, and other - - lation characteristics, with the highest rates across the lifeuiJ occurring among non-Hispanic American lndianr?Alaska Native and non-Hispanic White population groups. in 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.?3 Other population groups disproportionately impacted by suicide include middle?aged adults (whose rates increased 35% from 2000 to 2015, with Steep increases seen among both males and females aged 35-64 years?; lv'eterans and other military personnel {whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the ?rst time in decades)?; workers in certain occupational groupsmr? and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non?sexual minority peersu'? Suicides re?ect only a portion of the problem.'5 Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.15 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self- harm injuries, 27 who reported making a suicide attempt, and over 227' who reported seriously considering suicideFr?? Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal in?uences that interact with one another, often over time."5 The social ecological model?encompassing multiple levels of focus from the individual, relationship, community, and societal?is a useful framework for viewing and understanding suicide risk and protective factors identi?ed in the literature.? Risk and protective factors for suicide exist at each level. For example, risk factors include: - Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants - Relationship level: high con?ict or violent relationships, sense of isolation and lack of social support, family! loved one's history of suicide, ?nancial and work stress - Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications] - Societal level: availability of lethal - - of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness.?@ It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicidal? Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status.LS - Preventing Suicide: A Technical Package of Policy, Programs, and Practices 4am . .. Exposure to violence is associated with in creased risk of 4 depression, post-traumatic ress disorder (PTSD), aagimti Protective factors, or those in?uences that buffer against the risk for suicide, can also be found across the different levels of the social ecological model. Protective factors identi?ed in the literature include: effective coping and problem?solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means."5These protective factors can either counter a speci?c risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts.?25 Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.2E Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide attempts?? The effects of violence in childhood and adolescence can be observed decades later, including severe problems with ?nances, family, jobs, and stress factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove bene?cial in preventing suicide?? Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one's community,? school}2 family,33 caring adults,34r35and pro?social peers36 can enhance resilience and help reduce risk for suicide and other forms ofviolence. Preventing Suicide: it Technical Package of Policy, Programs, and Practices 3? The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals. families, and communities.?40 In an early study, Crosby and Slacks?H estimated that 7% of the U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide.They also estimated that for each suicide, 425 adults were exposed, or knew about the death.41 In a more recent study, in one state, Cerel et al?? found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide andfor having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long?term health and mental health consequences ranging from anger, guilt. and physical impairment, depending on the means and severity of the attempt.? Similarly, survivors of a loved one's suicide may experience ongoing pain and suffering including complicated grief,? stigma, depression, anxiety, post- traumatic stress disorder, and increased risk of suicidal ideation and Less discussed but no less important, are the ?nancial and occupational effects on those left behind.?7 The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work?loss costs alone.?7 Adjusting for potential under?reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self?directed violence to be approximately $93.5 billion in 201 3.43 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.43 The true economic costs are likely higher, as neither study included monetary ?gures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable.? While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public."5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta?analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found signi?cant preventive effects on suicide; cl meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or cl) evidence from at least one rigorous RCT or quasi- experimental design) evaluation study that found signi?cant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving bene?cial effects on multiple forms of violence; no evidence of harmful effects on speci?c outcomes or with particular subgroups; and feasibility of implementation in a U5. context ifthe program, policy, or practice has been evaluated in another country. Parenting Methnital ?drag! a?oat}. Progrmiull?radim Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help? seeking, stigma reduction, depression, connectedness). In terms ofthe strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors re?ect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often signi?cant heterogeneity among the programs, policies, or practices that fall within one approach or strategy in terms of the nature and quality of the available evidence. Not all programs, poli- cies. or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations.Tailoring programs and conducting more evaluations may be necessary to address different population groups.The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have bene?cial effects on risk or protective factors for suicide. Contextual and Cross-Cutting Themes One important feature ofthe package is the complementary and potentially synergistic impact ofthe strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box on page 12). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included ?rst, followed by those that might impact subsets of the population persons who have already made a suicide attempt]. ?ll? like most public health problems, . suicide is prewntable. I I A . Preventing Suicide: A Tech nical Package of Policy, Programs, and Practices 1 1 .tt Preventing Suicide Strategy Approach - Strengthen household ?nancial security Strengthen economic supportsli) Housing stabilization policies . Coverage of mental health conditions in health insurance policies - Reduce provider shortages in underserved areas - Safer suicide care through systems change Strengthen access and delivery of suicide care - Preschool enrichment with family engagement Create protectlve emnronments - Improved quality of child care through licensing - - ccreditation - Early childhood home visitation Promote connectedness - Parenting skill and family relationship approaches - Enhanced primary care Teach coping and - Behavioral parent training programs El problem-solving skills - Treatment to lessen harms of abuse and neglect exposure - Treatment to prevent problem behavior and later involvement in violence - Gatekeeper training Identify and support - Crisis intervention people at risk - Treatment for people at-risk of suicide - Treatment to prevent re-attempts Lessen harms and prevent - Postvention future risk Safe reporting and messaging about suicide It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social-emotional learning programs, an approach under the Teach Coping and Problem-Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative in?uences substance use] associated with suicide. Preventing Suicide: ATechnical Package of Policy, Program, and Practices The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden ofsuicide. Suicide ideation, thoughts, attempts. and deaths vary by gender, racefethnicity, age. occupation, and other important population Further, certain transition periods are also associated with higher rates ofsuicide transition from working into retirement, transition from active duty military status to civilian In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors eiiperienced.52 Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience.? identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or bene?cial effects on risk or protective factors for suicide is only the ?rst step. In practice, the effectiveness of the programs, policies and practices identi?ed in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the ?eld may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work.53'55 These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s pro?le of risk and protective factors with evidence?based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program materials, training and technical assistance) can also in?uence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sectorlnvolvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. Wanting Suicide: Todmical Package ofl'n?ch?mgramsmd We. 1? Strengthen Economic Supports for Families Rationale Studies from the US. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years old.55-5" Economic and ?nancial strain, such asjob loss, long periods of unemployment, reduced income, dif?culty covering medical, food, and housing expenses, and even the anticipation of such ?nancial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems.53 Buffering these risks can, therefore, potentially protect against suicide. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household ?nancial security and ensuring stability in housing during periods of economic stress. Strengthening household ?nancial security can potentially buffer the risk ofsuicide by providing individuals with the ?nancial means to lessen the stress and hardship associated with a job loss or other unanticipated ?nancial problems.The provision of unemployment bene?ts and other forms of temporary assistance, livable wages, medical bene?ts, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household ?nancial security. Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of?nancial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modi?cation programs, move-out planning, or ?nancial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates . Reductions in eviction rates - Reductions in emotional distress - Reductions in rates of suicide Preventing Suicide: I. Technical Package of Policy, Programs, and Fraction Evidence There is evidence suggesting that strengthening household ?nancial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal- Stote Unemployment insurance Program allows states to de?ne the maximum amount and duration of unemployment bene?ts that workers are entitled to receive after a job loss.59 An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment bene?ts {mean level: $7,990 per person in U.S. constant dollars}.The effects of unemployment bene?t programs were also consistent by sex and age group.59 Another US. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of Jlob, predicted suicide risk.? Together, these results suggest that not only should state unemployment bene?t programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen household ?nancial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical bene?ts, and other forms of family assistance) have also shown an impact on rates of suicide. A study by Flavin and Radcliff? examined the impact of statesF per capita spending on transfer payments, medical bene?ts, and family assistance {Temporary Assistance to Needy Families and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors leg, residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin and Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied.? At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of Preventing'Suidde: A Techni-l Package of Policy, Programs. arid Practices Evidence suggests . that stratigthenm?f? - household?nanci?l 7 housing can reduce suicide risk. '9 .I- no -. lulsi?.35- .17.. A assistance by $45 per year?" Although this was a correlational study, the results demonstrate the potential bene?ts of policies that reach particularly vulnerable individuals during periods of great need. More evaluation studies are needed to further understand the outcomes impacted by programs such as these..-. .. . "Housing stabilization policies. The Neighborhood was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration of the neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers ?nancial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults!33 Another study of data from 16 US. states participating in the Notionoi Vioient Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began] to 2010 [after it had peaked)? Most of these suicides occurred prior to the actual loss of the decedent's home. These ?ndings suggest that integrating suicide prevention resources, messaging, and referrals into ?nancial, foreclosure, and move-out planning and counseling services may help to prevent suicide. Preventing Suicide: it Technical Package of Policy, Programs, and Practices 3? Strengthen Access and Delivery of Suicide Care Rationale While most people with mental health problems do not attempt or die by suicide?19 and the level of risk conferred by different types of mental illness previous research indicates that mental illness is an important risk factor for suicide?? State-level suicide rates have also been found to be correlated with general mental health measures such as depressionfii-?g Findings from the National Comorbidity Survey indicate that relatively few people in the US with mental health disorders receive treatment for those conditions.m Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services.? Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention.S Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and ef?ciently deliver such care."2 Apart from treatment bene?ts, these approaches can also normalize help?seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity)?3 Bene?ts and services covered include such things as the number of visits, co- pays, deductibles, inpatienthutpatient services, prescription drugs, and hospitalizations. lfa state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. lfa state has a weaker parity law than the federal parity law le.g., includes coverage for some mental health conditions but not others}, then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which bene?ts and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Reduce provider shortages in underserved areas. Access to effective and state-of?the-art mental health care is largely dependent upon the training and the size ofthe mental health care workforce. Over 85 million Americans live in areas with an insuf?cient number of mental health providers; this shortage is particularly severe among low- income urban and rural communities.M There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering ?nancial incentives through existing state and federal programs loan repayment programs] and expanding the reach of health services through telephone, video and web?based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Preventing Suicide: ll Technical Package of Policy. ngrarns. and Practice 3? at a Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this isjust one piece of the puzzle. Care should also be delivered ef?ciently and effectively. More speci?cally, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identi?cation and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People tilt-Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes - Increased use of mental health services - Lower rates of treatment attrition . Reductions in depressive - Reductions in rates of suicide attempts - Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental heaith conditions in health insurance policies. The National Survey on Drug Use and Health is a nationally representative survey of the US. population that provides data on substance use, mental health conditions, and service utilization.? Using data from this survey, Harris, Carpenter, and Bad? found that 12 months after states enacted mental health parity llself-reported use of mental healthcare services signi?cantly increased. Moreover, subsequent research by Lan Ir xamined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, speci?cally, were associated with an approximate 5% reduction in suicide rates.This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year.59 Reduce provider shortages in underserved areas. One example ofa program to improve access to mental health care providers is the National Health Service Corps which offers ?nancial incentives to attract mental/behavioral health clinicians to underserved areas.? Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to the Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Telementai Health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance?9 TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. it can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance - zo Prevenungsuidsmremnial Access to health and behavioral health care services is critical for people at risk of suicide. abuse, and suicidal ideation and suicidef?g Further, Mohr and colleagues? conducted a meta?analysis examining the effect of delivered speci?cally via telephone and found that it signi?cantly reduced depressive in comparison to face-to-face also found that treatment attrition rates were signi?cantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy?? Thus, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered PerfectDepression Cares" the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly. the goal of the program was to redesign delivery of depression care to achieve ?breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, ef?ciency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow? up care system wide.?31 An examination of the impact of the program found that there was a dramatic and statistically signi?cant decrease in the rate of suicide between the baseline years, 1999 and 2000, die intervention years, 2002-2009. During this time period, the suicide rate fell by Further, among HMO members who received mental health specialty services, the suicide rate signi?cantly decreased over time from 1999 to 2010 {1 10.3 to 47.6 per 100,000 population; p<.04) with a mean of 36.2 per 100,000 over the for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 {p<.001 1.33 Preventing Suicide: A Technical Package of Policy, Programs, and Practices in" Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes.Em Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide."17 For example, rates of suicide are high among middle-aged adults who comprise 42.6% ofthe workforce?s,? among certain occupational groupsm'?; and among people in detention facilities jail, prison],EH5 to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are notim Similarly, modifying the characteristics ofthe physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition.39?94 Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. Reduce access to lethal means among persons at?risk of suicide. Means of suicide such as ?rearms, hanging! suffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a ?rearm in a suicide attempt will die from the injury}?5 Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes?? and 2) people tend not to substitute a different method when a highly lethal method is unavailable or dif?cult to accessf?ir?ig Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more dif?cult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at?risk of suicide: - intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help.1043 - Safe Storage Practices. Safe storage of medications, ?rearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing ?rearms locked in a secure place leg, in a gun safe or lock box), unloaded and separate from the ammunition; and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments leg. residential settings). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, ?nancial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolationist? alcohol use. Research studies In the United 1' States have found that greater alcohol availability is positively associated with alcohol?involved suicidesm'ms Policies to reduce excessive alcohol . use broadly include zoning to limit alcohol outlet .. - 1; locations and density, taxes on alcohol, and bans In on the sale ofalcohol for individuals under the I I I legal drinking age.? These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts.? il fair-:? . . Communit based olicies to reduce excessive . . . . m? wip?nwll'u I .. Potential Outcomes - Increases in safe storage of lethal means - Reductions in rates of suicide - Reductions in suicide attempts - Increases in help-seeking . Reductions in alcohol-related suicidedeaths 33 K: Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U5. and abroad, found associated reduced rates of suicideJ?m-W For example, after erecting a barrier on the Jacques?Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year."is Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to other jumping sites.mg Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased ?vefold-??11?? Another form of means reduction involves implementation of safe storage practices. In a case-control study of ?rearm?related events identi?ed from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, researchers found that storing ?rearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents.?0 Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices signi?cantly increased safe ?rearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one's own.101 Preventing Suicide: ATechnical Package of Policy, Programs, and Practices 0 Another program, the Emergency Department Counseling on Access to Lethal Means (ED trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 13 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al39 found that at post-test 76% (of the 55% of parents followed up, n:114l reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test (Le, all (100%) reported guns were currently locked up at post-test.Bg Organiaational policies and culture. Together for Life is a workplace program of the Montreal Police Force implemented to address suicide among of?cers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking.? Police suicides were tracked over 12 years and compared to rates in the control city of Quebec.The suicide rate in the intervention group decreased signi?cantly by to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city {29.0 per Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 1 1 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide.The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problems and instead sees them as larger service?wide problems impacting the whole community.? Using a time?series design to examine the impact of the program on various violence?related outcomes. researchers found that the program was associated with a 33% relative risk reduction in suicide."iThe program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found signi?cantly lower rates ofsuicide after the program was launched than before.? These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years.? Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff; standardized intake screening and risk assessment; provision of shared information between staff members (especially in transitioning or transferring of inmates); varying levels of observation; safe physical environment; emergency response protocols; noti?cation of suicidal behaviorisuicide through the chain of command; and critical incident stress debrie?ng and death review can potentially reduce suicide)? When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year.?3 Other similar programs have seen declines in suicide both in the United States and internationally.? ?4 Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use ofalcohol exist, several studies on alcohol outlet density and risk factors for suicide, such as interpersonal violence and social connectedness,?5??1 suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codesindicated that greater density of bars, speci?cally, igrelated to greater suicide and suicide attempts, particularly in rural areas.119 PmmtingSuidrle: Mammal Package . {uni r: imWr Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, lack of connectedness, aeqamong the chief causes feisuicidality?? Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others.?? Social connections can be formed within and between multiple levels of the social ecology,? for instance between individuals peers, neighbors, co-workers}, families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations?;123 Many ecological cross-sectional and longitudinal studie have examined the impact of aspects of social capital on depression depressive disorder, mental heal more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement], and improved mental Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversi 1 I nnectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs ofits members and provide collective primary prevention activities to the community as a whole.121 Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer narm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking I to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social in?uence of peers, these approaches can be used to shift group-level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in I community settings)? I I Community engagement activities Community engagement is an aspect of social capital??' Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, andehysical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, redUCed stress, and decreased depressive thereby reducing risk of suicide. Preventing Suicide: it Technical Package of Pulicy. Programs, and "3:11qu 0 Potential Outcomes Potential Outcomes Increases in healthy coping attitudes and behaviors Increases in referrals for youth in distress Increases in help?seeking behaviors Increases in positive perceptions of adult support If Promoting connectedness among individuals and within communities may protect against suicide. - 28 Preventing Suicide. ATechnIl Package of 'Policg, Programs, and Practices .A . . . 535,; Evidence Current evidence suggests a number of positive bene?ts of peer norm and community engagement activities. although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrengrh can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (6 metropolitan, 12 rural), researchers found that the program improved adaptive norms regarding suicide, connectedness to adults, and school engagement.36 Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths. particularly among those with a history of suicidal ideation, and the acceptability of help?seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders-?5 Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 18 million square feet) in 4 areas of the city. Researchers found signi?cant reductions in community residents*self?reported level of stress, wh-ieh is a risk factor for suicide, and engagement in more physical exercise. a protective factor for suicide. than residents in control vacant lot areas. There i-s?semeevidence for other cross-cutting bene?ts, including reductions in ?rearm assaults and vandalismm-?Eg Preventing Suicide: I. Technical Package of Policy, Programs, and Practices 3? Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, con?ict resolution, and critical thinking. Life skills are important in shieldingindividuals from suicidal behaviorsu?i Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,13g surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual leg. hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and ?nd solutions for problems has been characterized among suicide attempters?l Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social-emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. Sociahemotional learning programs focus on developing and strengthening communication and problem?solving skills, emotion regulation, con?ict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative in?uences substance use} associated with suicide.I26 These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work.?32 Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities.132 Programs are typically designed for parents or caregivers with children in a speci?c age range and can be self?directed or delivered to individual families or groups offamilies. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Speci?c program content typically varies by the age of the child but often has consistent themes of child development, parent?child communication and relationships, and youth?s interpersonal and problem-solving skills. Preventing Suicide: I. Technical Package of Policy, Programs, and Fraction 39F 3 Potential Outcomes - Reductions in suicide ideation - Reductions in suicide attempts - Reductions in suicide risk behaviors depression. anxiety, conduct problems, substance abuse] - Improvements in help-seeking behavior - Improvements in social competence and emotional regulation skills - Improvements in problem-solving and con?ict management skills Evidence Several social?emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse.133 Social-emotional learning programs. The Youth Aware ofMental Health Program (YAM) is a program developed for teenagers aged 14?16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety) and enhances their problem?solving skills for dealing with adverse life events, stress, school and other problems.134 In a cluster- randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were signi?cantly less likely to attempt suicide and have severe suicidal ideation at the 12+month follow?up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, ?ve attempted suicide in the YAM group compared to in the control group. Additionally, related to severe suicide ideation, in the YAM group, relative risk fell by Another example is the Good Behavior Game {686), which is a classroom-based program for eiementary school children aged 6-10.The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 686 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior.?35Two cohorts of youths participated in the program in 1985-86 and 198887 school years when they were in the ?rst and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide?related outcomes, an outcome evaluation of the 686 indicated that individuals in the ?rst cohort who were assigned to participate in 686 when they were in the ?rst grade reported half the adjusted odds of suicidal ideation and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting.The bene?cial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included.?35The GBG effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of 686 students, neither suicidal ideation nor suicide attempts were signi?cantly different between 686 and the control interventions.135 The researchers believed this may have been due to a lack of implementation ?delity, including less mentoring and monitoring of teachers. 636 was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the ?rst cohort of students. Results for the second cohort were generally smaller but in the desired direction.?5 Suicide: ?technical Package Policy. Program. and Practices 6 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years {in is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors in youth by improving protective factors such as responsive and positive parent-teacher?child interactions and relationships, emotion self-regulation and social competence (all protective factors for program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the iY program on reducing internalizing such as anxiety and depression, and child conduct problems.13mg The program is also associated with improved problem-solving and con?ict management; these skills were maintained at 1~year Additionally, the program demonstrated greater bene?ts in mother?rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.132 Additionally, Strengthening i0? i4 is a program that involves sessions for parents, youth, and families with the goal of improving parents? skills for disciplining, managing emotions and con?ict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide.142 Strengthening Families has been shown to signi?cantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families.142 Parenting and 1 family skills training approaches have shown promising impacts in pre ven ting key risk factors associated with suicide. is. hr . Preventing Suiclrle: ll Technical Package of Policy, Programs, and Practices l? Identify and Support People At?Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority Supporting peopie at-risk requires proactive case ?nding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible Internet?based services when appropriate] and engaging people in evidence? based care through such measures as collaborative treatment}, remain key challenges?m?t?r145 Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services.m Approaches The following approaches focus on identifying and supporting people at increased risk of suicide. Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk.?15 Crisis intervention These approaches provide support and referral services, typically by connecting a person in crisis {or a friend or family member of someone at-riski to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization?? Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Treatment for people at-risl-t of Suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with problem?solving and emotion regulation. Treatment usually takes place in a one-on-one or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed.Treatment that employs collaborative lie, between patient and therapist or care manager) andror integrated care linkage between primary care and behavioral health care) can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide Preventing Suicide: A Technical Package of Policy. Program. and Pram-u Treatment to prevent re-attempts. These approaches typically include follow?up contact and use diverse modalities te.g., home visits, mail, telephone, e?mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts?? Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one-on- one interpersonal therapy and/or group therapy. Approaches that engage and connect a?empte?to peers and providers are especially important because many attem pters do not present to aftercare; 12%?25% repttempt within a year, and of attempt survivors die by suicide within 1 to 5 years of their initial attemptm Potential Outcomes - Reductions in suicidal ideation - Reductions in suicide attempts - Reductions in suicide rates - Reductions in depression and feelings of hopelessness - Reductions in repttempts - Improvements in coping skills - Increases in treatment engagement and compliance with medications Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision oftreatrnent and support for these individuals can positively impact suicide and its associated risk factors. {Sate-keeper training Applied Suicide intervention Skilis Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8.: Kleinman?52 evaluated the training across the NotionaiSuicide Prevention Lifeline network of hotlines over the period 2008-2009. Using data from 1,41 0 suicidal individuals who called 17 Lifeline centers, the researchers found that callers who spoke with ASiST-trained 0 counselors were signi?cantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end oftheir call, compared to callers who spoke to non~ASiSTtrained counselors. Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training?? Gatekeeper training has also been a primary component of the GarretLee Smith (GL5) Suicide Prevention Program, which has been fundee?n 50 states and 50 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10-24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had signi?cantly lower youth suicide rates one year following the training implementation.?53This ?nding equates to a decrease of 1 suicide death per 100,000 amengyouth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had signi?cantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities [4.9 fewer attempts per 1000 More than 79,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Nationai Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half ofthe initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a signi?cant decrease in pain, hopelessness, and intent to die between initiation of the call [time 1) to follow?up (time 3i).155 Between time 2 {end ofthe call] to time 3, the effect remained for pain and hopelessness, but was not signi?cant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.155 Treatment for people at- risk of suicide. The improving Mood - Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. iMPACTfacilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase} by a depression care manager.156 The program has been shown to signi?cantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of relative to patients who received care as usual. Coiiaborative Assessment and Management of Suicidaiity (CAMS), is a therapeutic approach for suicide-speci?c assessment and treatment.The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-speci?c treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,?? in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual (intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed] found better treatment retention among the CAMS group and signi?cant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow-up.155 Preventing Suicide: A Technical Package of Fancy. Fmgrammnd Patties 3F Other examples include Dialectical Behavioral Therapy (DST) and Attachment?Based Family Therapy (ARI-T). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues.The components of DBT include individual therapy, group skills training, between?session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined?? a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositionalankiety.150 A randomized controlled trial ofABFTfound that suicidal adolescents assigned to ABFF experienced signi?cantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a signi?cantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%} and at 24 weeks [82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Solutions project uses a depression care liaison to link primary care and mental health services.The depression care liaison assesses and educates patients and follows-up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the ef?ciency of providing mental health services by bringing mental health care to the primary care setting, where most patients are ?rst detected and subsequently treated for many mental health conditions. An evaluation of found signi?cant decreases in depression severity scores among i096 of primary care patientsJ?' TiDESplso demonstrated 85% and 95% compliance with medication and follow-up visits, respectively.151 Treatment to prevent re ?attempts. Several strategies that aim to prevent re?attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Follow?up Visits is a program that involves a one?hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow?up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 13 months). Follow-up contacts are either conducted by phone or through home visits according to a speci?c timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in ?ve countries (Brazil, India, Sri Lanka, lran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with signi?cantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively).162 Another example of treatment to prevent re-attempts involves active follow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation.151 These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the ?rst 12 months post?discharge with some programs continuing contact for two or more years}. In a meta?analysis conducted by Inagaki et al151 interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce repttempts by approximately for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on r@ttempts has not yet been demonstrated.151 Also, because the number oftrials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death?laysuicide. 92mm Sul?de: Mammal mirage of Polity, Preamp. and Practices ized controlled trial ofthe post?crisis suicide prevention long?term follow?up contact approach, Motto; ?ostrowfound that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow?up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and Finally, Cognitive Behavior Therapy for Suicide Prevention is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related dif?culties} leading up to and following the event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family's problemgolving skills. A randomized controlled trial of found that iO?session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual."55 Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. ?1 6? . Preventing Suicide: it Tacit nical Package of Policy, Program, and Practices 39 ?9 Q. iSto 3" "by GEEKYI nag Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world.5 Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friendfpeer, family member, co-worker, or other close contact to suicide?? Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion??g'mg Approaches Some approaches that can be used to lessen harms and reduce future risk of suicide include postvention and safe reporting and messaging following a suicide. Postvention approaches are implemented after a suicide has taken place and may include debrie?ng sessions, counseling, and/or bereavement support groups for surviving friends and family membersfloved ones.These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief?? Safe reporting and messaging about suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline], and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagionm Potential Outcomes - Reductions in suicidal ideation - Reductions in suicide attempts - Reductions in rates of suicide - Reductions in distress - Improvements in reporting following suicide - Reductions in contagion effects related to suicide Preventing Suicide: ll. Technical Package of Policy, Programs, and Practices 3? 0 Evidence Current evidence suggests that postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postven lion. One example of a postvention program with evidence of impact on risk and protective factors for suicide is the StondBy Response Service {StondBy}. StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs.?2 In a study by Visser, Comans, and Scufl?ham,?2 StondBy clients were signi?cantly less likely to be at high risk for suicidality (suicide ideation and attempts} and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program {48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [veesus?pas?vgapproaches where survivors self-refer for Safe reparting and messaging about suicide- One way to ensure safe reporting and messaging about suicide is to encourage news media Jadhere to Recommendations for Reporting on Suicide org). The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time?eries design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide signi?cant reduction of81 suicides annuallyJE'g Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide.1M Reports of individual suicidal ideationpot accompanied by reports of suicide or suicide with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with signi?cant decreases in suicide rates in the time period immediately following such reports.1M Pastven tion and safe reporting and messaging can impact risk and protective factors. for suicide. 6 "aside: Ail?adlmical? arranq,rmgm,uamm 0 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy for Suicide Prevention,? the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, Iabor,justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems}, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Care. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness ofand garnering support for policies affecting individuals and families. The public health sector has been at the forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem?Solving Skills to prevent suicide from-happeai-ngin the ?rst place. These programs are often delivered in school and community settings, making education and non?governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can serve in an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Preventing Suicide: A Technical Package of Fancy. Fragrammnd hitting: 0 Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identity and SupportPeopieAt?Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families.These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful imple ation ofthis package. In this regard, all sectors can play an important and in?uential role in preventing suicide happening in the ?rst place and lessening the immediate and long?term harms ofsuicidal behavior by helping those in times of crisis get the services and support they need. 9 All sectors can .. play an important andin?uential role in preventing suicide. Images" Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent ofthe problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying de?nitions of suicidal ideation, behavior, and death that can make it dif?cult to consistently monitor speci?c outcomes across sectors and over time. For example, the manner in which deaths are classi?ed can change from one jurisdiction to another, and can change based on local medical andi'or medico?Iegal standards.4 CDC's uniform de?nitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems? Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's Notional Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of- death data from death certi?cates? is a state-based surveillance system [currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certi?cates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches.?S Data from state and local Child Death Review teams?? and Suicide Death Review Teams (which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modi?able risk factors for suicide. Preventing Suicide: ll Technical Package of Policy, Programs, and Practices 15? The Notional Electronic injury Surveillance System?All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause fails, poisoning, etc), age, race} ethnicity, sex, disposition (where the injured person goes when released from the Emergency In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-1 2 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose.?7The data are obtained from a national school?based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies?? The National Survey on Drug Use and Health (NSDUHFG is an annual survey ofthe civilian, non? institutionalized population aged 12 years and older. NSDUH provides both national and state?level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs}; mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk?? It is also important at all levels (local, state, and federal} to address gaps in responses, track progress of prevention efforts and evaluate the impact ofthose efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and longvterm outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on risk factors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes, Preventing Sul?de: .l-T?h?l?l Package of Policy. Programs, and Practices IE Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern.There are a number of barriers that have impeded progress, including, for example, stigma related to help?seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is and more is being done to prevent suicide than ever before, as evidenced by the work ofthe National Action Alliance for Suicide the release of the ?rst world report on suicide,5 and more timely surveillance data, to namejust a few examples. In an effort to continue pushing the ?eld and society further towards prevention, this technical package includes strategies and approaches that ideally would be used i - umprehensive, multi-level and multi?sectoral way. It includes strategies and approaches to prevent suicide .occurring in the ?rst place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds ofthe typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and ?nancial security}. While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. In keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be re?ned to re?ect the current state of the science. In closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible." ?Hope, help, and healing is possible.? WE as"? was?? v- -- iStock by a Fatty 1m References 11. 12. 13. 14. 15. 16. 18. 19. 20. U.S. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: 2012. National Action Alliance for Suicide Prevention. Action Alliance Priorities. 2017; http:f/ Frieden TH. Six components necessary for effective public health program implementation. Am Public Health. Crosby AE, Ortega L, Melanson C. Self?directed Violence Surveillance: Uniform De?nitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control;2011. World Health Organization. Suicide Prevention: A Global Imperative. Geneva, Switzerland: WHO Press; 2014. Centers for Disease Control and Prevention. Web-Based Injury Statistics Query and Reporting System (WISOARS). Atlanta, GA: National Center for Injury Prevention and Control. Available online: index.html. Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. KE, Canham-Chervak M, Black SA, Dada EO, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2002?8. Prev. Lineberry TW, O?Connor 55. Suicide in the US Army. Mayo Clinic Proceedings. McIntosh WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholow B. Suicide Rates by Occupational Group 17 States, 2012. MM WR Morb Mortal Wkly Rep. 2016;65l25lz?41?545. Han B, Crosby AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18v6??years in the United States. Compr 201 6:660 76?1 36. Kann L, McManus T, Harris WA, et al. Youth Risk Behavior Surveillance United States, 2015. MMWR CDC SurveillSurnm. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: Evidence from a national study. Am Public Health. 2001 :91 {81:1 226-1 281. Stone OM, Luo F, Ouyang L, Lippy C, Hertz MF, Crosby AE. Sexual orientation and suicide ideation, plans, attempts, and medically serious attempts: evidence from local Youth Risk Behavior Surveys, 2001 ~2009. Am Public Health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer J. Suicidal thoughts and behaviors among adults aged 24:18 years? United States, 2008-2009. MMWR Lipari R, Piscopo K, Kroutil LA, Kilmer Miller G. Suicidal thoughts and behavior among adults: Results from the 2014 National Survey on Drug Use and Health. NSDUH Data Review 201 Dahlberg LL, Krug EG. Violence-a global public health problem. In: Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization: 2002121. Owens D. Fatal and non-fatal repetition of self-harm: Systematic review. 2002;181 Olfson M, Gerhard T, Huang C, S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 20159211210 122?1 181. Bossarte RM, Karras E, Lu N, et al. Associations between the Dartment of Veterans Affairs? suicide prevention campaign and calls to related crisis lines. Public Health Rep (Washington, D.C.: 197:0. 2014;1 29(61516?525. Preventing Suicide: A technical Package of Pn?cy. Programsand Practices 0? 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. Chapman DP. Whit?eld CL. Felitti VJ, Dube SR. Edwards VJ. Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood..lAffect Disord. 2004;82l2lz212-225. Dube SR. Anda RF. Felitti VJ. Chapman DP. Williamson DF. Giles WH. Childhood abuse. household dysfunction. and the risk of attempted suicide throughout the life span: ?ndings from the Adverse Childhood Experiences Study. JAMA. 2001 28694130893096. Felitti VJ. Anda RF. Nordenberg D. et al. Relationship of childhood abuse and household dysfunction to many ofthe leading causes of death in adults:The Adverse Childhood Experiences (ACE) Study. Am Prev Med. Klomek AB. Sourander A. Gould M. The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research ?ndings. Can Leeb Lewis T. Zolotor AJ. A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. 201 World Health Organization. Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and Non?Partner Sexual Violence. Geneva: World Health Organization; 2013. Bellis MA. Hughes K. Leckenby N. et al. Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bull World Health Organ. 20149291541655. Haegerich TM. Dahlberg LL. Violence as a public health risk. American Journaloftifesh/le Medicine. 201 Wilkins N. Tsao B. Hertz M. Davis Fl. Klevens J. Connecting the dots: An overview of the links among multiple forms of violence. Atlanta, GA: National Center for injury Prevention and Control, Centers for Disease Control and Prevention; 2014. Hamby S. J.The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociology. New York. NY: Springer; 2013. Kleiman EM. Riskind JH. Schaefer KE. Weingarden H. The moderating role of social support on the relationship between impulsivityr and suicide risk. Crisis. Carter M. McGee R. Taylor 3. Williams 5. Health outcomes in adolescence: associations with family. friends and school engagement. JAdolesc. 2007;30( 1 1:5 1 ?62. Maimon D. Browning CR. Brooks-Gunn J. Collective ef?cacy. family attachment. and urban adolescent suicide attempts. Health Soc Capaldi DM. Knoble NB. Shortt JW. Kim HK. A Systematic Review of Risk Factors for Intimate Partner Violence. Partner Abuse. Losel F. Farrington DP. Direct protective and buffering protective factors in the development of youth violence. Prev Med. 201 2431.? Suppl 13:58?523. Wyman PA, Brown CH. LoMurray M. et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Public Health. Dunne EJ. JL. Dunne-Maxim K. eds. Suicide and its aftermath: Understanding and counseling the survivors. New York: Norton; 1987. Mishara BL. The lmpactofSuicide. New York: Springer;1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The Way Forward: Pathways to hope. recovery. and wellness with insights from lived experience. Washington. DC. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to Grief Trauma. and Distress After a Suicide: US. National Guidelines. Washington. D.C. Author; 2015. Crosby AE. Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States. 1994. Suicide Life Threat Behav. ?3 28. Cerel J. Maple M. De Venne A. Moore M. Flaherty C. Brown M. Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Health Rep (Washington. D.C.: Pum?m} Sumac: Mammal Package of Palm. Progmu. and Prawns 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 58. 59. 50. 61. 62. 63. 64. Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav. Mitchell AM, Kim Y, Prigerson HG, MortimersStephens M. Complicated grief in survivors of suicide. Crisis. 2004:2511 2? 18. Sudak H. Maxim K. Carpenter M. Suicide and stigma: a review of the literature and personal re?ections. Acad 36-1 42. Cerel J. McIntosh JL, Neimeyer RA, Maple M, Marshall D.The Continuum of"Survivorship": De?nitional Issues in the Aftermath of Suicide. Suicide Life Threat Behav. 201 -600. Florence C, Simon T, Haegerich T, Luo F, Zhou C. Estimated lifetime medical and work?loss costs of fatal injuries?United States. 2013. MM WR Mora Mortai Wkiy Rep. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr., Silverman MM. Suicide and Suicidal Attempts in the United States: Costs and Policy implications. Suicide Life Threat Behav. Silverman MM, Maris prevention of suicidal behaviors: An overview. Suicide Life Threat Behav. 21. Substance Abuse and Mental Health Services Administration. Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016; Desai RA, Dausey DJ. Rosenheck RA. Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am 2005;162i2iz31 1n318. Turecki G. Epigenetics and suicidal behavior research pathways. Am Prev Med. Edwards JumpervThurman P, Plested BA, Getting ER. Swanson L. Community readiness: Research to practice. Journai of Community Hawkins JD, Catalano RF, Kuklinski MR. Communities that care. Encyciopedia ofCriminoiogy and Criminai Justice: Springer; 2014:393-403. Plested BA, Edwards RW, Jumper-Thurman P. Community readiness: A handbook for successful change. Fort Collins. CO: Tri-Ethnic Center for Prevention Research. 2006. Luo F, Florence CS, Quispe?Agnoli M, Ouyang L, Crosby AE. Impact of business cycles on US suicide rates, 1928?2007. Am Public Heaith. 201 Fowler KA, Gladden RM, Magi KJ, Barnes J. Frazier L. increase in suicides associated with home eviction and foreclosure during the US housing crisis: ?ndings from 16 National Violent Death Reporting System States, 2005?2010.Am Pubiic Heaith- 2015;10512ir311-316. Stack S, Wasserman l. Economic strain and suicide risk: a qualitative analysis. Suicide Life Threat Behav. 200?;37l1 ):103? 1 12. Cylus J, Glymour MM, Avendano M. Do generous unemployment benefit programs reduce suicide rates? A state ?xed- effect analysis covering 1968-2008. Am JEpidemioi. Classen TJ, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass?layoffs and unemployment duration. Heaith Econ. 2012;21 Flavin P. Radcliff B. Public policies and suicide rates in the American states. Sociai indicators Research. 209. US. Department of Housing and Urban Development. Neighborhood Stabilization Program. hudexchangeinfoiprogramsinspi, 2017. Houle JN. Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. AmJPubiic Heaith. Arsenauit-Lapierre G, Kim C.Turecki (3. diagnoses in 3275 suicides: a meta-analysis. BMC 2004.437. Pmmting Suicide: Todmical Package ofi'n?cy. Pragmatism! imam 0 65. 66. 67. 68. 69. 70. 71. 7'2. 73. 24. 75. 76. 2'8. 29. BO. 81. 32. 83. B4. 85. 86. BB. 89. Harris EC, Barraclough B. Suicide as an outcome for mental disorders. A meta-analysis. 1992;120:205? 228. Tyrer P, Reed GM, Crawford MJ. Classi?cation, assessment, prevalence, and effect of personality disorder. Lancet. 201 7-726. Harris EC, Barraclough B. Excess mortality of mental disorder. 1998;173:1163. Mark TL, Shern DL, Bagalman JE, Cao Z. Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. 2007. Lang M.?lhe impact of mental health insurance laws on state suicide rates. Health Econ. 2013;22l1 1:23?83. Wang PS, Demler O, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Am Public Health. 2002;92l1 1:92-98. Cunningham PJ. Beyond parity: primary care physicians? perspectives on access to mental health care. Health All (Millwood). 2009;281312w490-501. Coffey CE. Building a system of perfect dEpression care in behavioral health. it Con'lmJr OualPatientSaf. The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 HR 1424, 1 10*h Congress. 2008. US. Department of Health and Human Services Health Resources and Services Administrations. Designated Health Professional Shortage Areas National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author; 2014. Harris KM, Carpenter C, Bao Y.The effects of state parity laws on the use of mental health care. Med Care. Health Resources 8: Services Administration. National Health Service Corps. 201?; U.S. Department of Health and Human Services Health Resources and Services Administrations. National Health Service Corp Clinician Retention: A Story of Dedication and Commitment. 20161 Hailey D, Roine R, Dhinmaa A. The effectiveness of telemental health applications: a review. The ConJ Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice. 2008;1 Coffey CE. Pursuing perfect depression care. Serv. Coffey CE, Coffey MJ, Ahmedani BK. An update on perfect depression care. 2013;641:41396. Coffey M, Coffey C, Ahmedani BK. Suicide in a health maintenance organization population. JAMA 2015:7213i:294?296. Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep. Toosi M. Labor force projections to 2024: The labor force is growing, but slowly. Washington, D.C.: Bureau of Labor Statistics; Labor Review. 2015:1-33. Noonan ME. Mortality in State Prisons, 2001?201 4 Bureau ofJustice Statistical Tables. 201 6:250150lDecember]. Knox KL, Pflanz S,Ta cott (3W, et al. The US Air Force suicide prevention program: implications for public health policy. Am Public Health. 2010;100i1 212452-2463. National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Blueprint for Workplace Suicide Prevention. Washington, DC. Author; 2015. Runyan CW. Becker A, Brandspigel 5, Barber C, Trudeau A, Novins D. Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emerg Med. - 52 Pum?ngSuldde: Mammal htlrigeolPollqr. Progmmn?raaim 90. 91. 92. 93. 94. 95. 96. 98. 99. 100. 101. 102. 103. 104. 105. 106. 10?. 103. 109. 110. 111. 112. Miller M, Warren M, Hemenway D, Azrael D. Firearms and suicide in US cities. in} Prev. 201 5:21 {e1 }:e1 16-1 19. Crosby AE, Espitia~Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Alcohol: Science, Policy and Public Health. 2013:190?193. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: a gender-strati?ed analysis of the National Violent Death Reporting System. in] Prev. 201 3:1 911 ):38-43. Beautrais AL, Gibb SJ, Fergusson DM, Horwood LJ, Larkin GL. Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Journal of Correctional Health Care. 2015;21 [31:222?242. Elnour AA, Harrison J. Lethality of suicide methods. in} Prev. 2000-1 411139-45. Simon 0R, Swarm AC, Powell KE, Potter LB, Kresnow MJ, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001 :32{1 Deisenhammer EA, CM, Strauss R, Kemmler G, Hinterhuber H, Weiss duration of the suicidal process: how much time is left for intervention between consideration and accomplishment ofa suicide attempt?J Clin Hawton K. Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. Yip P, Caine E, YousufS, Chang S-S, Wu K, Chen Means restriction for suicide prevention. Lancet. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health. 201 3:1 3(1 Rowhani?Rahbar A, Simonetti JA, Rivara FP. Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Hayes LM. Suicide prevention in correctional facilities: reflections and next steps. 194. Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off? premise and on?premise alcohol outlet density. Addiction. 2015;1 Escobedo LG, Ortiz M.The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Xuan Z, NaimiTS, Kaplan M5, at al. Alcohol Policies and Suicide: A Review of the Literature. Alcohol Clin Exp Res. 201 Cherpitel CJ, Borges GL6, Wilcox HC. Acute alcohol use and suicidal behavior: A review ofthe literature. Alcoholism: Clinical and EXperimental Research. 2004286 Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: A systematic review and meta-analysis. 2015:2[1 11994?1001. Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation ofa bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: A case study. 2001 Grossman DC, Mueller BA, Rledy C, et al. Gun storage practices and risk of youth suicide and unintentional ?rearm injuries. JAMA. Mishara BL, Martin N. Effects ofa comprehensive police suicide prevention program. Crisis. 2012:33l3121 62-168. Knox KL, Litts DA,Ta cott GW, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ. 2003:322l7428izi 376. Preventing Suicide: I Tedi nlcal Fading! of Fancy. Programmed Practices 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 12?. 12B. 129. 130. 131. 132. 133. 134. 135. 136. Hayes LM. Prison Suicide: An Overview and a Guide to Prevention. The Prison Journal. 1995:25i4iz431-456. Barker E, Kolves K, De Leo D. Management of Suicidal and SelfrHarming Behaviors in Prisons: Systematic Literature Review of Evidence?Based Activities. Archives afSuicide Research. 201 4:1 881222?240. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol-related damage. 1. The distribution of consumption model.JStud Alcohol. Gruenewald PJ, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res. Lipton H, Gruenewald P. The spatial dynamics of violence and alcohol outlets. Stud Alcohol. 95. Lippy C, DeGue S. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Violence, .9.- Abuse. Johnson FW, Gruenewald PJ, Remer LG. Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res. Durkheim E. Suicide: A study in sociology (JA Spaulding 8: G. Simpson, trans). Glencoe, it: Free Pressi'Original work published 189?). 1892/1951. Centers for Disease Control and Prevention. Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. 2009: Available at: Muennig P, Cohen AK, Palmer A, Zhu W.The relationship between ?ve different measures of structural social capital, medical examination outcomes. and mortality. Soc Sci Med. Beyer KM, Layde PM, Hamberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other fernicides? Violence Against Women. Whitley Fl, McKenzie K. Social capital and review of the literature. Harv Rev 2005:1 De Silva MJ, McKenzie K, Harpham T, Huttly SR. Social capital and mental illness: a systematic review. Epidemiol Community Health. 2005:59i812619-622. Wyman PA. Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med. 2014:47i3 Suppl 21:5251?256. Centers for Disease Control and Prevention. Principles of community engagement. Committee on Community Engagement. 19971 Branas CC, Cheney RA, MacDonald JM, Tam VW. Jackson TD, Ten Have TR. A difference-in-differences analysis of health, safety, and greening vacant urban space. Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald JM. Urban blight remediation as a cost-bene?cial solution to ?rearm violence. Am Public Health. 2016:106l12it2158?2164. Bandura A. Social foundations ofthought and action: A social cognitive theory. Prentice-Hall, Inc: 1936. Pollock LR, Williams JM. Problem?solving in suicide attempters. Med. 2004:34i1 1:163?1 62. Herman KC, Borden LA, Reinke WM, WebstenStratton C. The Impact of the Incredible Years Parent, Child, and Teacher Training Programs on Children?s Co?Occurring lnternalizing Cl. 201 Knox MS, Burkhart K, Hunter KE. ACT against violence parents raising safe kids program: Effects on maltreatment? related parenting behaviors and beliefs. Journal of Family issues. 201 0. Wasserman D, Hoven CW, Wasserman C, et al. School?based suicide prevention programmes: The SEYLE clusterv randomised, controlled trial. Lancet. Wilcox HC, Kellam 56, Brown CH, et aI.The impact of two universal randomized ?rst? and second?grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend. 2008:95 Suppl 1:560?23. Kellam 56, Brown CH, Poduska JM, et al. Effects of a universal classroom behavior management program in ?rst and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend. 2008:95 Suppl 1255-528. Pomona] Suldde: Mammal Package bl Palm. Program. and Prawns 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. Webster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high-risk schoolsJournalofChild and 2008:49l511421-488. Webster-Stratton CH, Reid MJ, Beauchaine T. Combining parent and child training for young children with ADI-ID. Journal of Clinical Child 3; Adolescent 201 91 -203. Reid Ml, Webster-Stratton C. Hammond M. Follow~up of children who received the Incredible Years intervention for oppositional?de?ant disorder: Maintenance and prediction of 2?year outcome. BehaviorTherapy. 2003:34141:421?491. Webster-Stratton C, Hammond M. Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin 1992:6511193-109. Webster-Stratton C, Reid MJ, Hammond M. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 2001 30131233302. Spoth RL, Guyll M, Day 5X. Universal family?focused interventions in alcohol?use disorder prevention: cost- effectiveness and cost~bene?t analyses of two interventions. JStua' Alcohol. 20026391219228. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics: 2016. Jobes DA. The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence-based clinical approach to suicidal risk. Suicide Life Threat Behav. 201 3. Wilcox HC, Wyman PA. Suicide Prevention Strategies for Improving Population Health. Child Adolesc Clin Am. 2016;25:219233. Isaac M, Elias B, Katz LY, et al. Gatekeeper training as a preventative intervention for suicide: a systematic review. CanJ Gould MS, Munfakh JL, Kleinman M, Lake AM. National suicide prevention lifeline: enhancing mental health care for suicidal individuals and other people in crisis. Suicide Life Threat Behav. 201 Gilbody S. Bower P, Fletcher J, Richards D. Sutton AJ. Collaborative care for depression: a cumulative meta?analysis and review of longer?term outcomes. Arch lntMea?. 2006:16612112314?2321 . ArcherJ, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews. 2012:10{Art No D0065 25]. Bruce ML, Ten Have TR, Reynolds CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA. 2004:291 {91:1081-1091. Inagaki M. Kawashima Y, Kawanishi C. et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. AffectDisord. 2015;125:6648. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. 201 Walrath C, Garraza LG, Reid H, Goldston DB. McKeon R. Impact ofthe Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. 201 310561986993. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 2015:72l1 11:1 143?1149. Gould MS, Kalafat J, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav. 2002:32131338-352. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ. Unutzer J, Tang L, Oishi S, et al. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc. 2006;54i10121550?1556. Preventing Package ofl'n?cy. mom 55F 158. 159. 160. 161. 162. 163. 1 64. 165. 166. 167. 168. 169. 170. 171. 172. 173. 1 74. 176. 177. Comtois KA, Jobes DA, S. O?Connor 5, et al. Collaborative assessment and management ofsuicidality feasibility trial for next-day appointment services. Depress Anxiety. Linehan MM, Comtois KA, Murray AM, et randomized controlled trial and follow?up ofdialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Diamond GS. Wintersteen MB, Brown GK, et al. Attachmentabased family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Chiid Adolesc 31. Rubenstein Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Systems, .8: Health. 13. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in ?ve countries. Bali World Heaith Organ. Motto JA, Bostrom A6. A randomized controlled trial of suicide prevention. 833. Hassanian-Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. 2011;198i4iz309-316. 1Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life ThreatBehav. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention ofsuicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects of suicide bereavement on mental health and suicide risk. Lancet Etzersdorfer E, Sonneck C1. Preventing suicide by in?uencing mass-media reporting:The Viennese experience, 1980- 1996. Arch Suicide Res. NiederkrotenthalerT, Sonneck G. Assessing the impact of media guidelines for reporting on suicides In Austria: Interrupted time series analysis. Aust 200?;41 (51:419?428. Szumilas M, Kutcher S. Post-suicide intervention programs: A systematic review. Can Pubiic Bohanna l, Wang X. Media guidelines forthe responsible reporting of suicide: a review of effectiveness. Crisis. VisserVS, Comans TA, Scuffham PA. Evaluation Of The Effectiveness Of A Community?Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 2014;42i1 1:19-23. Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role ofan active postvention model. Suicide Life Threat Behav. 200838? 1:30-34. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. BMJ. 2010;341:6841. Centers for Disease Control and Prevention. National Violent Death Reporting System. Atlanta, GA: Centers for Disease Control and Prevention, National Center for injury Prevention and Control; 2017. Available online: The National Center for the Review 8: Prevention ofChild Deaths. U.S. Child Death Review Programs. Centers for Disease Control and Prevention, Brener ND, Kann L, et al. Methodology ofthe Youth Risk Behavior Surveillance System--2013. MMWR Recomm Rep. - 56 Prm?ngSuldde: Mammal hckagenfh?q.Progmu.an?mtlm L. Tihages?? W. a . Appendix: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Practice or Policy Suicide Suicide Attempts or ldeation Other Risk! Protective Factors for Suicide Lead Sectors? Strengthening household ?nancial security Government Strengthen a (local, state, economic hemp oyment ene tprograms Federal] supports ?3th inmme SUPPOTFS ?l Businessl La bor Housing stabilization policies Gwemmem . (local, state, Nerghborhood Stabilization Program 1/ Federal} Coverage of mental health conditions in health insurance policies Mental Health Parity La ws Government Strengthen Reduce provider shortages in underserued areas (local, state, Federal and National Health Service Corps delivery {If Telemental Health (TMl-li of Healthcare surcide care Safer suicide care through systems change Social Services Henry Ford Perfect Depression Care (Pre-cursor to Zero Suicide) Reduce access to lethal means among persons at-risk Government intervening at suicide hot spots (local, state) Safe storagepractices 1/ Public Health Emergency Department Counseling on Access to lethali'vleans (so CALM) Healthcare Create Organizational policies and culture Busmessl labor protective Togethe, for We Justice US Air Force Suicide Prevention Program Govern ment . . . . 1, (local, state, Correctional saicicle prevention Federal) Community-based policies to reduce excessive alcohol use Government (local, state) Alcohol outlet density .f ii" Business/labor Peer norm programs Public Health Promote Sources of Strength Education connectedness Community engagement activities Public Health . Govern ment Greening vacant urban spaces v? (local) l*This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing speci?c activities. Preventing Suicide: ATechnical Package of Policy, Program. and Practices Best Available Evidence . . her Risk! n1, Suicide 0t . 1 Strategy Practice or Policy Attempts or Protective Lead Sectors . Factors for Ideatrnn . . Surcrde Social?emotional learning programs Youth Aware ofMental Health Teach coping :eaEhBPLogramG Education and 00 dirt-or time solving skills Parenting skill and family relationship approaches Public Health The incredible Years Strengthening Families 30? Education Gatekeeper training Applied Suicide intervention bliC Health Skills Training Garret Lee Smith Federal v, Health Care Grant Program Crisis intervention Public Health National Suicide v? Prevention Lifeline 50'3?! Services Treatment for people at risk of suicide improving Mood Promoting Access v, Identify and to Collaborative Treatment sup port people Collaborative Assessment and Healthcare at-risk Management ofSurcrdalrty (CAMS) Social Services Dialectical Behavioral Therapy (DST) v? A ttachment-Based Family Therapy 1/ (A EFT) Translating initiatives for Depression into Effective Solutions project i Treatment to prevent re-attempts ED Brief in tervention with v? alth are Follow?up Visits Active follow?up contact approaches v? Social Services EST for Suicide Prevention Postvention ea care Lessen harms StandBy Response Service and prevent fe . 'd . future risk a reporting an message a out surcr public Health Media Guidelines s/ Media l*This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy: there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing speci?c activities. Preventing Suicide: A Padrage of Policy, Programs. and "will For more information To learn more about preventing child abuse and neglect, call 1-800-CDC-INFO or visit violence prevention pages at NatiOnal Center for Injury Prevention and Control Division of Violence Prevention 3' DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgments We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access?to mental?hem and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening?to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention's priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called seif?o'irecteo? vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, Melanson, 2011). Self- directed violence may be suicio?oi or non-suicide! in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by seif?directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the US (Centers for Disease Control and Prevention, 2016). Overall suicide rates increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2014, the rates for these groups were 17.8 and 16.4 per 100,000 population, respectively (Centers for Disease Control and Prevention, 2016). Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 48% from 1999 to 2014, with steep increases seen among both males and females aged 45-64 years; (Curtin et al., 2016); Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in et al., 2012; Lineberry 8: O'Connor); workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; McIntosh et al., 2016); and lesbian, gay, bisexual, and/or queer youth, who experience increased suicidal ideation and behavior compared to their heterosexual counterparts (Kann et al., 2016; Russell Joyner, 2001). Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8i. Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one's history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office ofthe Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) increases the risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts (Bossarte et al., 2014; D. P. Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, 8: Gould, 2010; Leeb, Lewis, 8: Zolotor, 2011; World Health Organization, 2013). Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other 8 instability, increases the risk for suicide and suicide attempts several fold (Bellis et al., 2014; Dube et al., 2001). The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide (Haegerich 8v: Dahlberg, 2011; Hamby 2013; Wilkins, Tsao, Hertz, Davis, Klevens, 2014). Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community (Kleiman, Riskind, Schaefer, 8; Weingarden, 2012), school (Carter, McGee, Taylor, 8: Williams, 2007), family (Maimon, Browning, Brooks-Gunn, 2010), caring adults (Capaldi, Knoble, Shortt, 8: Kim, 2012; Losel Farrington, 2012), and pro-social peers (Wyman et al., 2010) enhances resilience to suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). in an early study, Crosby and Sacks (2002) estimated that 2% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et al (2016) found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental heaith consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (A. L. Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2003). Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegerich, Luo, 8; Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.3 billion in estimated lifetime medical and work-loss costs alone (Florence et al., 2015). Adjusting for potential under-reporting ofsuicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, 81 Silverman, 2016). The overwhelming burden of these costs were from lost productivity over the life course, with the 9 average cost per suicide being over $1.3 million {Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (US. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman E: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public le.g., business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014L Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available 10 evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cu tting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports - Strengthen household financial security I Housing stabilization policies Strengthen access and delivery of suicide u. Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments Reduce access to lethal means among persons at- risk of suicide - Organizational policies and culture - Community-based policies to reduce excessive alcohol use Peer norm programs Community engagement activities Promote connectedness Teach coping and problem-solving skills Social?emotional learning programs I Parenting skill and family relationship approaches Identify and support people at risk - Gatekeeper training Crisis Intervention I Sereeningeneielreatment for people at riskgf suicide I Treatment to prevent re~attempts Lessen harms and prevent future risk - Postvention . Safe reporting following a suicide 11 It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teach Coping and Problem- Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences le.g., substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status}. In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced {Turecki, 2014). Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Getting, Swanson, 2000; Hawkins, Catalano, 81 Kuklinski, 2014; Plested, Edwards, 8: Jumper-Thurman, 2006). These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation le.g., availability of program 12 materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts]. The role of various sectors in the implementation of a strategy' or approach in preventing suicide is described further in the section on Sector invoivement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide (Stack 8a Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 14 Potential Outcomes I Reductions in foreclosure rates a Reductions in eviction rates - Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal-State Unemployment lnsuronce Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss (Cylus, Glymour, 8c Avendano, 2014). An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: $7,990 per person in US. constant dollars; Cylus et al., 2014). The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5?14 weeks, 15-26 weeks, and greater than 26 weeks}, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk {Classen Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. In terms of lives saved, Flavin 8L Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year {Flavin 81 Radcliff, 2009). Although this was a correlational study, 15 the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Nationai Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslv evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analvsis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014}. Another studv of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent's home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, 8: Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {Arsenault-Lapierre, Kim, 81 Turecki, 2004; E. C. Harris 8: Barraclough, 1997; Tyrer, Reed, El Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 8: Barraclough, 1998; World Health Organization, 2014). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8i. Cao, 2007}. Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention (World Health Organization, 2014). Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care Coffey, 2007). Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. if a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health 17 Comment IAI: Cross?check the outcomes with the evidence presented providers; this shortage is particularly severe among low?income urban and rural communities (US. Department of Health and Human Services Health Resources and Services Administrations, 20153}. There are a number ofways to increase the number and distribution of practicing mental and make sure everything is health providers in underserved areas including offering financial incentives through existing state and federal programs expanding telemental health services. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services critical for people at risk of suicide; however this is just one piece of the puzzle. Care being accessed?mustshould also be defivereo? efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments {see ldentf?r and Support People ?it-Risk, p.31), continuity of care, and continuous quality improvement. Care that is patient?centered and promotes equity for all patients is also of critical importance {Nationai Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014). Potential butcomeg Increases in access to mental health services Increase in utilization of mental health services Reductions in of mental illnesses and suicidality - Reductions in rates of suicide attempts - Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the deiivery of care can red risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The Nationai Survey of Drug - Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006) found that 12 months after states enacted mentoi heoith parity tows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013] examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated 18 with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year (Lang, 2013). Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps which offers financial incentives to attract mental/behavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps lU.5. Department of Health and Human Services Health Resources and Services Administrations, 2015b}. Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide greventionTelementol health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used variety of settings leg. outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deathsmeetal?ity among other outcomes {Hailey, Roine, E: IlL?ihinmaa,r 2008). 19 Comment Suggest dropping this piece. It mostly speaks to primary care physicians and the funding is precarious. Further, Mohr and colleagues (2008) conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone- administered compared to patients receiving face-to-face therapy. Thus, may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford healthca re system, which is a large health maintenance organization in the state of Michigan pioneered the Perfect Depression Core program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve ?breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8: Ahmedani, 2013). Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 {110.3 to 47.6 per 100,000; p<.04) with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 (p<.001) (M. Coffey, Coffey, Ahmedani, 2015i 20 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes [Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide (Dahlberg 8i Krug, 2002; U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults {who comprise 42.6% of the workforce; Toosi, 2015}; among certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016), and among people in detention facilities leg. jail, prison}, to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition {Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; A.E. Crosby, Espitia-Hardeman, Ortega, 8t. Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Byck, Teplin, 2015}. Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. II Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001), and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 Comment IAI: Program doesn't seem to he the right word here. Are you referring to residential care facilities or something along those lines?I Comment Be sure to do a cross- walk betvreen the outcomes listed here and the evidence described in the next section. 0 intervening otSur?cr?a?e Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013}. Sofe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable Individuals from easy access to lethal means. Such practices may include education and counseling around ll storing firearms?docked in a secure place leg, in a gun safe or lock box], unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts i i i lRowhani-Rahbar, Simonetti, SI Rivara, 2015; Runyon et al., 2015]. I Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential Eerogramssettings?. Such policies and cultural values encourage leadership from;I the top down and may promote prosocial behavior le.g., asking for help), skill building, positive social norms, assessment, referral and access to helping services leg. mental health, substance abuse treatment, ?nancial counseling], and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation} {Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015}. Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides [Escobedo Ortiz, 2002,- Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, takes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one?third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004Potential butcomes} Increases in safe storage of lethal means Reductions in suicide attempts Reductions in suicide deaths 22 Increases in help-seeking Reductions in alcohol-related suicide deaths Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015}. For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, 31 Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement ofsuicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani?Rahbar et al., 2016). Another program, The Emergency Department Counseiing on Access to Lethoi Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2016} found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components Iwere designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifving suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help seeking {Mishara 81L Martin, 2012}. Police suicides were tracked over 12 vears and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantlv by to a rate of 6.4 suicides per 100,000 population per vear compared to an 11% increase in the control citv {29.0 per 100,000,- Mishara 81. Martin, 2012). Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policv and education initiatives and was designed to change the culture of the Au Force surrounding suicide. The program uses leaders as role modeis and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates everv suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solelv as medical problem and instead sees them as larger servicea wide problems impacting the whole communityI (Knox, Litts, Talcott, Feig, Caine, 2003] Using a time series design to examine the impact of the program on various violence?related outcomes researchers found that the program was associated with a 33% relatIve risk reduction in suicide {Knox et al., 2003}.The program was also associated with relative rIsk reductions in related outcomes including moderate and severe familv violence {30% and 54%, respectivelv) homicide and accidental death {Knox et al., 2003}. A longitudinal assessment of the program over the period 1981 to 2008 [15 vears before the 1992 launch of the program and 11 years post launch) found significantlv lower rates of suicide after the program was launched than before (Knox et al., 2010}. These effects were sustained over time, except In 2004, which the authors found was associated with less rigorous implementation in that 1,iear than in the other years [Knox et al. ,.2010] tsainingfer construction workees Comment IAI: I don?t think the evidence is strong enough to include this example. I found a studvr in addition to the economic study you cite below, but the relative reduction among construction workers was not significant. Just mentioned this to Jim and he doesn?t think it is strong enough to Include. feethe Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests that_organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behaviorfsuicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year (Hayes, 1995}. Other sirniiar programs have seen declines in suicide both in the United States and internationally (Barker, K?lves, 81 De Leo, 2014]. Community?based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density, speci?cally, suggest that measures to reduce alcohol outlet density can potentially reduce alcoholvinvolved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8: Homer, 2009}. 25 Comment IAI: Non-significant reduction - they believe the study was :underpowered. JL Comment IAI: Martin G, Swannell S, Milne-r A, Gullestrup] [2016} Mates in Construction Suicide Prevention Program: A Five Year Review. 1 Community Medicine and Health Education, 5:455. Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951}. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002], for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, 8; Zhu, 2013]. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest the pattern is towards a positive association between social capital measured by social trust, community/neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009). Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 26 enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors I Increases in referrals for youth in distressed I Increases help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (E metropolitan, 12 rural), Wyman et al. (2010) found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010}. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism {Branas et al., 2011). 27 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004}. Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on develooing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, &Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors Ii.e., depression, anxiety, conduct problems, substance abuse) 0 Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8: Hunter, 2010). Social emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wasserman et al., 2014). In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts amongthe YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and relative risk fell by 49.6% (Wasserman et al., 2014}. Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008]. Two cohorts of youths participated in the program in 1985-86 and 1985?87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation 29 and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions (Wilcox et al., 2008}. The researchers believed this may have been due to a lack of implementation fidelity. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence {all protective factors for suicide) (Herman et al., 2011). The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8: Hammond, 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionally, Strengthening Families 10?14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guyll, 8: Day, 2002). Strengthening has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 30 Comment IAI: Not sure the rationale fully captures the treatment pieces; Lseems a bit heavy on access. ?-[Comment IAI: Made edits below Comment [Als This sentence seemed Rationald In order to decrease suicide, care of, and attention to,peepleeHneFeased?risk vulnerable populations is If problematic because we were saying we need to pay attention to people at increased risk because they have higher . rates ofsuicidal behavior?that appeared previously; lv'eterans and active duty military personnel; indlviduals who are have been Lcircular 50changed_ victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial (.[Commem This no?, falls under the and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; g" Curtin et al., 2015; Kann et al., 2016; Lineberry O'Connor; Russell 8r. Joyner, 2001]. Supporting these Comment lAi: irevised the wording here, so that it is more clear that we are sag?Jr. identify and Support People At-Risk necessary, as these iadieiciuais?groups tend to experience suicidal behavior at higher than average rates. Fhesegughvulnerable populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have previously attempted suicide vuinerabiegroups requires proactive case response, crigis interventionguggd 5 ill areas and not ?disadvantaged people". In Comment lthoughtitsounded odd talking about economically disadvantaged evidence-based treatmentE alongwith access to,? aneL aetentiere in,? mentai health semaes} Finding [If effeetiveggtimaj ways of identifying at?risk customizing services to . make them more accessible leg, internet-based services when appropriate} and engaging peopleed in ii mention to low income. I care leg. through 51.1520 measures tLgatmentl, remain key challenges. Fer?eieamplergsimply improving or expanding services does not guarantee that those services will be used by thesepep-plgmost in need-ef?shem, nor will it necessarily increase the number of people who follow referrals or treatment. For example, some pPeople living in Morash?gface social and economic issues that c_a_n?adve rselyJ-?I affect their ability to access supportive services. Approaches The following approaches focus on identifying and supporting people at increased risk. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at?risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. 31 Comment IAI: Need to do a cross?wall: with the evldence presented for each approach to make sure all relevant outcomes are included here {and not just Treatment for people at?risk of suicide can include various forms of delivered by the ones that were there before the i licensed providers to help individuals with mental health problems and other suicide risk factors with problem-solving and emotion regulation. Treatment usually takes place in a one on one or group format betvveen patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/or integrated care le.g., linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk (Archer et al., 2012; Bruce et al., 2004; Gilbody, Bower, revision]. Fletcher, Richards, 8; Sutton, 2006}. In Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities leg, home visits, mail, telephone, e-mail] to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one? on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296-2596 reatternpt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (Inagaki et al., 2015Potential bottomed . RedUCtions in suicide attempts Reductions in suicide deaths Increases in identification of individuals at-risk for suicidal behavior Increases in at?risk individuals in treatment Increases in community members trained to identify at-risk individuals Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. I- Gatekeeper training. Applied Suicide intervention Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 81 Kleinman (2013] evaluated the 32 training across the Notionoi Suicide Prevention Lifeiine network of hotlines over the period 2008- 2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, the researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors without training in were significantly more likely to feel depressed, suicidal, more overwhelmed, and less hopeful by the end of their call to the hotline compared to those with training in ASLST. Counselors trained in ASLST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASLST did not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013). Gatekeeper training has also been a primary component of the Garret Lee Smith (GL5) Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation (Walrath, Garraza, Reid, Goldston, McKeon, 2015). This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities (4.9 fewer attempts per 1000 youths; Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015). More than 79,000 suicide attempts may have been prevented during the period examined, following implementation of the GL5 program. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 ofthose completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1] to follow-up (time Between time 2 (end of the call} to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die (Gould, Kalafat, Harrismunfakh, Kleinman, 2007). 33 Treatment for people atnrisk of suicide. The improving Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase) by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up (Hunkeler et al., 2006; Unutzer et al., 2006] relative to patients who received care as usual. Collaborative Assessment and Management of Suicidality (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies lJobes, 2012), in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual (intake with a or nurse practitioner followed by 1?11 visits with a case manager and medication as needed} found better treatment retention among the CAMS group and significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow-up (Comtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy (ABFT). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment {23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined (Linehan et al., 2006}. ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up 34 {Comment Attempts? Drshould this than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage participants reported no suicidal ideation in the week prior to assessment at 53]? "salf'harm"? 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%} and at 24 weeks The Veterans Affairs Transiating initiatives for Depression into Effective Soiutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the i efficiency of providing mental health services by bringing mental health care to the primary care Il setting, where most patients are first detected and subsequently treated for many mental health 5 conditions. Ag evaluationstoday of the?TiDES signi?cant decreases in 5' 5 i (82.1% vs. 46.2%] (Diamond et al., 2010]. i depression severity scores among 70% of primary care patients (Rubenstein et al., 2010}. ?0055 also demonstrated 85% and 95% compliance with medication and follow-Lip visits, respectively (Rubenstein et al., 2010}. II Treatment to prevent reattempts. Several strategies that aim to prevent reattempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief] intervention with Follow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and ttem ts distress, risk and protective( factors, alternatives to self?harm, and referral options, over 18 months (at 1, 2, 4, 11 weeks and 4, 6, 12, 13 months]. Follow-up contacts are either conducted by phone or through home visits according to a speci?c time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, lran, and China] found that a brief intervention combined with 9 follow-up visits over Lit-months was associated with significantly fewer deaths from suicide relative to a treatment-as?usual group versus respectively} (Fleischmann et al., 2008]. Another example of treatment to prevent re-attempts involves active foiiow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years]. In a metaranalysis conducted by Inagaki et al. (2015], interventions to prEVent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by for up to 12 months post-discharge; however, the effects of these approximately 12% approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the 35 number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. In a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts lHassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CST-5P also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial of found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual {Brown et al., 2005). 36 Lessen Harms and Prevent Future Risk [Rationalei -. - are suicide every year in the United States and throughout the world. Risk of suicide and suicide risk factors Comment Need to modify the eaoalsehass been shown to increase among thosepeople who have lost a friendipgr, family member, co-worker, or other contact to suicide (Pitman, Osborn, King, at Erlangsen, 2014). Care and attention to_the bereaved is thefe?lggf high im_p_qrtance. Despite the best of intentions, mMedia and others responding to suicide may inadyeeteetly?add to this risk-ameng?thelaeteaved. For example, research suggests thateExposure to sensationalized or otherwise uninformed reporting Fegasdiegon suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to uhatjg?noww?suicide contagion [Etzersdorfer 8: Sonneck, 1998; Niederkrotenthaler Sonneck, 2007}. the evidence is still being built in this area, particularly with regard to the impact of policy and practi_i;_es on suicide and suicide attempts, faking effeetive measures to care for thisthe bereaved population through such means as postvention interventions counseling. support groups and debriefing sessions) and safe reporting on suicide may reduce risk of suicide: Approaches EFhese?aee?a?nueaber?of?ome approaches that can be used to lesson harms and reduce future risk of suicide includgi-ng for the bereaved,? and safe reporting following a suicide. I Postvention approaches are implemented after a suicide has taken place and may include complicated grief (Szumilas 3-: Kutcher, 2011). contagionl debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicid contagion. F'herefore, re5ponsib e and safe reporting may help prevent suicide and suicid rationale so that it speaks to the primary approaches included below perhaps starting with the second sentence and ,expanding on that a bit. Comment IM: This introductory I . statement needs to be modified now that this section only includes postvention and I ksafe reporting followi?g a suicide. i i JL 1 Comment Suggest flashing this out a bit so that the reader has a good sense of what is meant by "safe reporting?. 3? Potential putcomesi .- . . . . I- Improvements in messaging following suicide - Reductions in re-attempts - Reductions in contagion effects related to suicide Ewdence Eurrent evidence suggests that therapeutie? treasmeatsand ether-appreac?hes?ke lessening harm through postvention and safe reporting can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, Stond?v Response Service i5tond3y}, provides clients with iace-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, 8i Scuffham, 2014). In a study by Visser et al. StondBy clients were significantly less likely to be at high risk for suicidality than a suicide bereaved comparison group who had not had contact with the StondBy program {48% and 64% respectively}. Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors self?refer for services} (Cerel 8: Campbell, 2008}. Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendations for Reporting on Suicide Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources leg, hotline], and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion. The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant redudion of 31 suicides annually {Niederkrotenthaler 8: Sonneck, 2007). Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition] have 38 J..- and safe reporting are included. Comment IAI: Statement needs to be modified Comment IAI: Outcome section needs to be modi?ed now that only postvention i harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010). Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports (Niederkrotenthaler et al., 2010}. Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact ofsuicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth?serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, mghealth sector [including insurersL-a-ad-providers, and health systems], and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non- governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. 39 The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teoch Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Creote Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies leg, criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 40 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andfor medicolegal standards (AE. Crosby, Ortega, et al., 2011). uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems A.E. Crosby, Ortega, et al., 2011). Surveillance systems exist at the federal, state, and local levels. it is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System (Ni/55) and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of?death data from death certificates. is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. Data from state and local Child Death Review teams and Suicide Death Review Teams {which are in a few states] offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department). in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade 41 students and is a key resource in monitoring health?risk behaviors among youth, including whether youth Comment IAI: Last sentence added per suggestion of EDIE. You can cite D?v?P's new have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by i Strategic Vision. a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2313}. The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Heoith is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state?level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs}; mental health [past year mental illness, co?occurring illnesses]; and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk. It is also important at all levels (local, state, and federal] to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long? term outcomes are an important part of program evaluation. The evidencevbase for suicide prevention has advanced greatiy over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum], as opposed to merely examining their effectiveness on risk factors. More research is also needed to examine the effectiveness of primary prevention strategies {before risk occurs} and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic efFECts within a comprehensive prevention approach. w, there are also many potential opportunities to build and strengthen partnerships across program areas leg, violence prevention, substance abuse prevention] to evaluate the impact of different approaches on multiple butcomedConclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis - at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help?seeking, 42 mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 43 References Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., . . . Coventry, P. (2012). Collaborative care for depression and anxiety problems. The Cochrane Library. ArsenauIt-Lapierre, G., Kim, (2., 8L Turecki, G. (2004). diagnoses in 3225 suicides: a meta-analysis. BMC 4, K. E., Canham-Chervak, M., Black, S. A., Dada, E. 0., Millikan, A. M., S: Jones, E. H. (2012). Mental health risk factors for suicides in the US Army, 2007?8. injury Prevention, 18(6), 405-412. Ba ndura, A. (1986). Saciaifoundatians of thought and action: A social cognitive theory: Prentice-Hall, Inc. Barker, E., K?lves, K., 8i. De Leo, D. (2014). Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research, 18(3), 227-240. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi:10.1080f00048620902823714 Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukol, B. (2014). Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the Wario' Health Organization, 92(9), 641-655. Beyer, K. M., Layde, P. M., Harnberger, L. K., 8; Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49-64. doi:10.1127f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 3: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i?-ill. Bossarte, R. M., Karras, E., Lu, N., Tu, R., Stephens, B., Draper, J., 8: Kemp, J. E. (2014). Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Heaith Rep, 129(6), 516?525. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. El Ten Have, T. R. (2011]. A difference? in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol, 1 74(11), 1296-1306. doi:10.1093lajeikwr273 Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. R., Hollander, J. E., Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?5?0. Bruce, M. L, Ten Have, T. R., Reynolds C. F., Katz, 1. Schulberg, H. C., Mulsant, B. H., . . .Alexopoulos, G. S. (2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291(9), 1081?1091. Capalcli, D. M., Knoble, N. B., Shortt, J. W., Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., E: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full version?a.pdf Centers for Disease Control and Prevention. (2016). Web?Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: 44 Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. Aicohoi: Science, Poiicy and Pubiic Heaith, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 5513), 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directed Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. {2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and iife-threatening behavior, 32(3), 321-328. Cunningham, P. i. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff 28(3), w490-501. Curtin, S. C., Warner, M., St Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 45 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 81'. Levy, S. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Doran, C. M., Ling, R., Gullestrup, J., Swannell, 5., 8: Milner, A. (2016). The Impact ofa Suicide Prevention Strategy on Reducing the Economic Cost of Suicide in the New South Wales Construction Industry. Crisis, 37(2), 121-129. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., Giles, W. H. (2001). Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA, 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1937}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8t G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 1897). Edwards, R. W., JumpervThurman, P., Plested, B. A., Getting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 2918,02. Escobeclo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 689694. Etzersdorfer, E., 81 Sonneck, G. (1993). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Felitti, V.J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks,J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245-258. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., 8.: Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi:50042? 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., 8t Zhou, C. (2015). Estimated lifetime medical and workeloss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 6488), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 8: Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005?2010. Am Public Health, 105(2), 311-316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi:10.2105lAJPH2013301608 Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 46 Gilbodv, 5., Bower, P., Fletcher, J., Richards, 0., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternai medicine, 166(21), 2314-2321. Godoy Garraza, L., Walrath, C., Goldston, D. 6., Reid, H., St McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. (2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. doi:10.1111isltb.12049 Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. Grossman, D. C., Mueller, 6. A., Riedv, C., Dowd, M. 0., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Haegerich, T. M., 8: Dahlberg, L. L. (2011). Violence as a public health risk. American Jaurnai aftifestyie Medicine, 1559822611409127. Hailey, D., Roine, R., at Ohinmaa, A. (2008). The effectiveness of telemental health applications: a review. The Canadian Joarnai of 53(11), 769278. Hambv, 5., J. (2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse Briefs in Sacioiogy. New York, NY: Springer. Han, B., Crosbv, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., 8; Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 126-186. Harris, E. C., at Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205228. Harris, E. C., 8: Barraclough, B. (1993). Excess mortality of mental disorder. Bri 173, 11?53. Harris, K. M., Carpenter, C., 3: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8; Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital~treated selfnpoisoning. British ioarnai of 198(4), 309-316. Hawkins, J. 0., Catalano, R. F., 8-1 Kuklinski, M. R. (2014). Communities that ca re Encyciopedia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Haves, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 75(4), 431-456. Haves, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int;r Law 138-194. doi:10.1016ij.ijlp.2013.04.010 Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring lnternalizing Q, 26(3), 189-201. Houle, J. N., St Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am i Pubiic Heaith, 104(6), 1073-1079. 47 Hunkeler, E. M., Katon, W., Tang, L., Williams, J. Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(7536), 259-263. Inagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad201412.048 lobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8.: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12], 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. (2015). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 1-174. doi:10.15585immwr.ssE550631 Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. S., Wang, W., Toyinbo, F., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55?528. Kessler, R. (2., Demler, 0., Frank, R. (5., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515?2523. doi:10.1056lN Elk/153043266 Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 223?279. doi:10.1027l0227? 591013000136 Klomek, A. E., Sourander, A., 8; Gould, NI. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 13?6. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), ?3u88. Leeb, R. T., Lewis, T., 81 Zolotor, A. J. (2011). A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of [lifestyle Medicine, 454-468. Lineberry, T. W., O'Connor, 5. S. Suicide in the US Army. Mayo Clinic Proceedings, 871-378. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 757-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. 48 Maimon, D., Browning, C. 8., 8; Brooks-Gunn, i. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Sherri, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8c Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. {2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention :8 Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. G., 8: Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8: Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243-253. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21? 41. Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 leson, M., Gerhard, T., Huang, C., S., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Perron, S., Burrows, 5., Fournier, M., Perron, P. A., Duellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 Pirkis, 1., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 8: Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. 49 Pitman, A., Osborn, 0., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Plested, B. A., Edwards, R. W., 31. Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rittenhouse, D. R., Fryer, G. E., Phillips, R. L., Miyoshi, T., Nielsen, C., Goodman, D. C., Grumbach, K. (2008). Impact of Title VII training programs on community health center staffing and National Health Service Corps participation. The Annais of Famiiy Medicine, 392-405. Rowhani-Rahba r, A, Simonetti, J. A, 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiai Rev, 38(1), 111-124. Rubenstein, L. V., Chaney, E. F., Ober, 5., Felker, B., Sherman, S. E., Lanto, A., at Vivell, S. (2010}. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Systems, Heaith, 28(2), 91. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emeryr Med, 17(1), 8-14. Russell, 5. T., Joyner, K. {2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Jaurnai of Pubiic Heaith, 91(8), 1276-1281. Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, G. 84 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352?362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and iife? threatening behavior, 25(1), 10?2 1. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 4959. Spoth, R. L., Guyll, M., Day, 5. X. (2002). Universal family+focused interventions in alcoholsuse disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 228. Stack, 5., 8t Wasserman, l. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and We threatening behavior, 37(1), 103-112. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. FL, 8: Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. .i Correct Heaith Care, 21(3), 222u242. Sudak, H., Maxim, K., 8! Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. doi:10.1176fappi.ap.32.2.136 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Pubiic Heaith, 102(1), 18-29. Toosi, M. (2015). Labor force projections to 2024: The iaborforce is growing, but siowiy. Washington, Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, G. M., 8: Crawford, M. J. {2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2016. Retrieved from 50 U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Clinician Retention: A Story of Dedication and Commitment. Retrieved from U.S. Office of the Surgeon General, at National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, DC. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. 5., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journaiof Community 42(1), 19-28. Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2015}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(sszz), Webster?Stratton, C., Hammond, M. Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consuit Ciin 65(1), 93?109. Webster?Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283302. Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible veers teacher and child training programs in highsrisk schools. Journal of child and 49(5), 471?488. Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., St. Anthony, J. C. (2008). The impact of two universal randomized first? and classroom interventions on voung adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?23. Wilkins, N., Tsao, B., Hertz, M., Davis, H., 8L Klevens, J. (2014). Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. World Health Organization. (2013). Giobal and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and Non?Partner Sexual Violence. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurrav, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8; Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, environments Unemployment benefit programs 1/ state, Fed eral) Strengthen economic Other income supports Business/labor . . . . . . Housmg stabilization policies Government (local, The National Neighborhood Stabilization state, Fed erall Program Coverage of mental health conditions in health insurance policies Government (local, Mental Health Parity Laws 1/ 1/ state, Federal) Strengthen access and Reduce provider shortages in underserved areas Healthcare delivery of National Health Service Corps suicide care Telemental health (TMHJ Social SEFVFCES Safer suicide care through systems change Henry Ford Perfect Depression Care (Pre- cursor to Zero Suicide) Reduce access to lethal means among persons at?risk Government (local, intervening at suicide hot spots state) v" Create Safe storage practices Public Health . protective 50 Best Available Evidence Organizational policies and culture Business/Labor Together for Life 1? Justice US Air Force Suicide Prevention Program 1/ v? Government {local, state, Federal) Community-based policies to reduce excessive alcohol use Government (local, state) Alcohol outlet density v" Businessilabor Peer norm programs Public Health Promote Sources of Strength 1/ Education connectedness Community engagement activities Public Health Greening vacant urban spaces Government local} Social emotional learning programs Public Health Youth Aware of Mental Health Program Education Teach coping . .x and problem- Good Behavior Game solving skills Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening 10-14 Best Available Evi - Identify and at-risk support people Gatekeeper training Public Health . . . . . . . Healthcare Apphed Samde in terventian Training Crisis Intervention Public Health National Suicide Prevention Lifeiine Social Services Treatment for people at risk of suicide Healthcare improving Mood Promoting Access to Coiiahorative Treatment Social Services Caiiahorative Assessment and Management Justice ofSuicidaiity (CAMS) Diaiecticai Behaviorai Therapy (DST) Attachment-Based Family Therapy Transiating initiatives for Depression into Effective Soiutions project Treatment to prevent re-attempts Heaith ca re ED Brief intervention with Foiiow-ap Visits Social services Active foiiow?ap contact approaches 1/ CBT for Suicide Prevention 1? 52 Best Available Evidence Postvention Health ca re Lessen harms Stono?By Response Service and prevent future risk Safe reporting following a suicide Public Health Media Guidelines 1/ Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgments We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access?to mental?hem and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening?to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention's priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called seif?o'irecteo? vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, Melanson, 2011). Self- directed violence may be suicio?oi or non-suicide! in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by seif?directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the US (Centers for Disease Control and Prevention, 2016). Overall suicide rates increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2014, the rates for these groups were 17.8 and 16.4 per 100,000 population, respectively (Centers for Disease Control and Prevention, 2016). Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 48% from 1999 to 2014, with steep increases seen among both males and females aged 45-64 years; (Curtin et al., 2016); Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in et al., 2012; Lineberry 8: O'Connor); workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; McIntosh et al., 2016); and lesbian, gay, bisexual, and/or queer youth, who experience increased suicidal ideation and behavior compared to their heterosexual counterparts (Kann et al., 2016; Russell Joyner, 2001). Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8i. Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one's history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office ofthe Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) increases the risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts (Bossarte et al., 2014; D. P. Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, 8: Gould, 2010; Leeb, Lewis, 8: Zolotor, 2011; World Health Organization, 2013). Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other 8 instability, increases the risk for suicide and suicide attempts several fold (Bellis et al., 2014; Dube et al., 2001). The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide (Haegerich 8v: Dahlberg, 2011; Hamby 2013; Wilkins, Tsao, Hertz, Davis, Klevens, 2014). Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community (Kleiman, Riskind, Schaefer, 8; Weingarden, 2012), school (Carter, McGee, Taylor, 8: Williams, 2007), family (Maimon, Browning, Brooks-Gunn, 2010), caring adults (Capaldi, Knoble, Shortt, 8: Kim, 2012; Losel Farrington, 2012), and pro-social peers (Wyman et al., 2010) enhances resilience to suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). in an early study, Crosby and Sacks (2002) estimated that 2% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et al (2016) found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental heaith consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (A. L. Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2003). Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegerich, Luo, 8; Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.3 billion in estimated lifetime medical and work-loss costs alone (Florence et al., 2015). Adjusting for potential under-reporting ofsuicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, 81 Silverman, 2016). The overwhelming burden of these costs were from lost productivity over the life course, with the 9 average cost per suicide being over $1.3 million {Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (US. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman E: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public le.g., business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014L Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available 10 evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cu tting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports - Strengthen household financial security I Housing stabilization policies Strengthen access and delivery of suicide u. Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments Reduce access to lethal means among persons at- risk of suicide - Organizational policies and culture - Community-based policies to reduce excessive alcohol use Peer norm programs Community engagement activities Promote connectedness Teach coping and problem-solving skills Social?emotional learning programs I Parenting skill and family relationship approaches Identify and support people at risk - Gatekeeper training Crisis Intervention I Sereeningeneielreatment for people at riskgf suicide I Treatment to prevent re~attempts Lessen harms and prevent future risk - Postvention . Safe reporting following a suicide 11 It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teach Coping and Problem- Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences le.g., substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status}. In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced {Turecki, 2014). Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Getting, Swanson, 2000; Hawkins, Catalano, 81 Kuklinski, 2014; Plested, Edwards, 8: Jumper-Thurman, 2006). These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation le.g., availability of program 12 materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. It also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts]. The role of various sectors in the implementation of a strategy' or approach in preventing suicide is described further in the section on Sector invoivement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide (Stack 8a Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 14 Potential Outcomes I Reductions in foreclosure rates a Reductions in eviction rates - Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal-State Unemployment lnsuronce Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss (Cylus, Glymour, 8c Avendano, 2014). An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: $7,990 per person in US. constant dollars; Cylus et al., 2014). The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5?14 weeks, 15-26 weeks, and greater than 26 weeks}, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk {Classen Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. In terms of lives saved, Flavin 8L Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year {Flavin 81 Radcliff, 2009). Although this was a correlational study, 15 the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Nationai Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslv evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analvsis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014}. Another studv of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent's home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, 8: Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {Arsenault-Lapierre, Kim, 81 Turecki, 2004; E. C. Harris 8: Barraclough, 1997; Tyrer, Reed, El Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 8: Barraclough, 1998; World Health Organization, 2014). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8i. Cao, 2007}. Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention (World Health Organization, 2014). Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care Coffey, 2007). Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. if a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health 17 Comment IAI: Cross?check the outcomes with the evidence presented providers; this shortage is particularly severe among low?income urban and rural communities (US. Department of Health and Human Services Health Resources and Services Administrations, 20153}. There are a number ofways to increase the number and distribution of practicing mental and make sure everything is health providers in underserved areas including offering financial incentives through existing state and federal programs expanding telemental health services. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services critical for people at risk of suicide; however this is just one piece of the puzzle. Care being accessed?mustshould also be defivereo? efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments {see ldentf?r and Support People ?it-Risk, p.31), continuity of care, and continuous quality improvement. Care that is patient?centered and promotes equity for all patients is also of critical importance {Nationai Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014). Potential butcomeg Increases in access to mental health services Increase in utilization of mental health services Reductions in of mental illnesses and suicidality - Reductions in rates of suicide attempts - Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the deiivery of care can red risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The Nationai Survey of Drug - Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006) found that 12 months after states enacted mentoi heoith parity tows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013] examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated 18 with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year (Lang, 2013). Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the National Health Service Corps which offers financial incentives to attract mental/behavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps lU.5. Department of Health and Human Services Health Resources and Services Administrations, 2015b}. Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide greventionTelementol health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used variety of settings leg. outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deathsmeetal?ity among other outcomes {Hailey, Roine, E: IlL?ihinmaa,r 2008). 19 Comment Suggest dropping this piece. It mostly speaks to primary care physicians and the funding is precarious. Further, Mohr and colleagues (2008) conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone- administered compared to patients receiving face-to-face therapy. Thus, may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford healthca re system, which is a large health maintenance organization in the state of Michigan pioneered the Perfect Depression Core program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve ?breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8: Ahmedani, 2013). Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 {110.3 to 47.6 per 100,000; p<.04) with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 (p<.001) (M. Coffey, Coffey, Ahmedani, 2015i 20 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes [Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide (Dahlberg 8i Krug, 2002; U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults {who comprise 42.6% of the workforce; Toosi, 2015}; among certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016), and among people in detention facilities leg. jail, prison}, to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition {Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; A.E. Crosby, Espitia-Hardeman, Ortega, 8t. Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Byck, Teplin, 2015}. Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. II Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001), and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place program doesn't seem to relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), ,l bl?- the right here. ME YOU referring i to residential care facilities or something railway tracks, and Isolated locations such as parks. Efforts to prevent sumde at these along museums? locations include erectln barriers or limitin access to revent 'um in and installin 3 g, 3 Comment Be sure to do a cross- walk between the outcomes listed here i signs and telephones to encourage individuals who are considering suicide, to seek help g' I and the evidence described in the next ll section. I {Cox et al., 2013}. Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable Individuals from easy access to lethal means. Such practices may include education and counseling around ll storing firearms?docked in a secure place leg, in a gun safe or lock box], unloaded and i separate from the ammunition--and keeping medicines in a locked cabinet or other Il secure location away from people who may be at risk or who have made prior attempts il lRowhani-Rahbar, Simonetti, S: Rivara, 2015; Runyan et al., 2015]. I Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential Eeregramssettings?. Su_ch policies and cultural values encourage leadership from;I the top down and may promote prosocial behavior le.g., asking for help), skill building, positive social norms, assessment, referral and access to helping services leg. mental health, substance abuse treatment, ?nancial counseling}, and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation} (Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015}. Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides [Escobedo 8i Ortiz, 2002,- Giesbrecht et al., 2015}. Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts {CherpiteL Borges, E: Wilcox, 2004Potential butcomed a Increases in safe storage of lethal means - Bed eaten-untamedsuicide 22 - Reductions in suicide attempts I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths Eviden ce The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015}. For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold lBeautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethal Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2015] found that at post- test 76% (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components Iwere designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifving suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help seeking {Mishara 81L Martin, 2012}. Police suicides were tracked over 12 vears and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantlv by to a rate of 6.4 suicides per 100,000 population per vear compared to an 11% increase in the control citv {29.0 per 100,000,- Mishara 81. Martin, 2012). Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policv and education initiatives and was designed to change the culture of the Au Force surrounding suicide. The program uses leaders as role modeis and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates everv suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solelv as medical problem and instead sees them as larger servicea wide problems impacting the whole communityI (Knox, Litts, Talcott, Feig, Caine, 2003] Using a time series design to examine the impact of the program on various violence?related outcomes researchers found that the program was associated with a 33% relatIve risk reduction in suicide {Knox et al., 2003}.The program was also associated with relative rIsk reductions in related outcomes including moderate and severe familv violence {30% and 54%, respectivelv) homicide and accidental death {Knox et al., 2003}. A longitudinal assessment of the program over the period 1981 to 2008 [15 vears before the 1992 launch of the program and 11 years post launch) found significantlv lower rates of suicide after the program was launched than before (Knox et al., 2010}. These effects were sustained over time, except In 2004, which the authors found was associated with less rigorous implementation in that 1,iear than in the other years [Knox et al. ,.2010] tsainingfer construction workees Comment IAI: I don?t think the evidence is strong enough to include this example. I found a studvr in addition to the economic study you cite below, but the relative reduction among construction workers was not significant. Just mentioned this to Jim and he doesn?t think it is strong enough to Include. feethe Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests that_organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behaviorfsuicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year (Hayes, 1995}. Other similar programs have seen declines in suicide both in the United States and internationally (Barker, K?lves, 81 De Leo, 2014]. Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studios on alcohol outlet density, speci?cally, suggest that measures to reduce alcohol outlet density can potentially reduce alcoholvinvolved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is inversely related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 3; Remer, 2009). 25 Comment IAI: Non-significant reduction - they believe the study was :underpowered. Comment Martin G, Swannell S, Milner A, Gullestrup] [2016} Mates in Construction Suicide Prevention Program: A Five Year Review. Community Medicine and Health Education, 5:455. JL Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951}. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002], for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, 8; Zhu, 2013]. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest the pattern is towards a positive association between social capital measured by social trust, community/neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009). Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 26 enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors I Increases in referrals for youth in distressed I Increases help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (E metropolitan, 12 rural), Wyman et al. (2010) found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010}. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism {Branas et al., 2011). 27 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004}. Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on develooing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, &Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors Ii.e., depression, anxiety, conduct problems, substance abuse) 0 Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8: Hunter, 2010). Social emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wasserman et al., 2014). In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts amongthe YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and relative risk fell by 49.6% (Wasserman et al., 2014}. Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008]. Two cohorts of youths participated in the program in 1985-86 and 1985?87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation 29 and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions (Wilcox et al., 2008}. The researchers believed this may have been due to a lack of implementation fidelity. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence {all protective factors for suicide) (Herman et al., 2011). The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8: Hammond, 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionally, Strengthening Families 10?14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guyll, 8: Day, 2002). Strengthening has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 30 Comment IAI: Not sure the rationale fully captures the treatment pieces; Lseems a bit heavy on access. ?-[Comment IAI: Made edits below Comment [Als This sentence seemed Rationald In order to decrease suicide, care of, and attention to,peepleeHneFeased?risk vulnerable populations is If problematic because we were saying we need to pay attention to people at increased risk because they have higher . rates ofsuicidal behavior?that appeared previously; lv'eterans and active duty military personnel; indlviduals who are have been Lcircular 50changed_ victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial (.[Commem This no?, falls under the and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; g" Curtin et al., 2015; Kann et al., 2016; Lineberry O'Connor; Russell 8r. Joyner, 2001]. Supporting these Comment lAi: irevised the wording here, so that it is more clear that we are sag?Jr. identify and Support People At-Risk necessary, as these iadieiciuais?groups tend to experience suicidal behavior at higher than average rates. Fhesegughvulnerable populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have previously attempted suicide vuinerabiegroups requires proactive case response, crigis interventionguggd 5 ill areas and not ?disadvantaged people". In Comment lthoughtitsounded odd talking about economically disadvantaged evidence-based treatmentE alongwith access to,? aneL aetentiere in,? mentai health semaes} Finding [If effeetiveggtimaj ways of identifying at?risk customizing services to . make them more accessible leg, internet-based services when appropriate} and engaging peopleed in ii mention to low income. I care leg. through 51.1520 measures tLgatmentl, remain key challenges. Fer?eieamplergsimply improving or expanding services does not guarantee that those services will be used by thesepep-plgmost in need-ef?shem, nor will it necessarily increase the number of people who follow referrals or treatment. For example, some pPeople living in Morash?gface social and economic issues that c_a_n?adve rselyJ-?I affect their ability to access supportive services. Approaches The following approaches focus on identifying and supporting people at increased risk. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at?risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. 31 Comment IAI: Need to do a cross-walk with the evidence presented for each .i approach to make sure all relevant I . outcomes are included here {and not Just Treatment for people at?risk of suicide can include various forms of delivered by the ones that were there before the licensed providers to help individuals with mental health problems and other suicide risk factors with problemvsolving and emotion regulation. Treatment usually takes place in a one on one or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative lie, between patient and therapist or care manager) and/or integrated care leg, linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in El revision], therapy and decreasing suicide risk (Archer et al., 2012; Bruce et al., 2004; Gilbody, Bower, i i i i Fletcher, Richards, 8: Sutton, 2006). I- Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities leg, home visits, mail, teiephone, e-mail] to engage recent suicide attempt survivors in continued treatment to prevent re?attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one? on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not i present to aftercare; 1296-2596 reatternpt within a year, and 396-996 of attempt survivors die by i suicide within 1 to 5 years of their initial attempt (inagaki et al,, 2015] Potential bottomed . Reductions in suicide attempts Reductions in suicide deaths 0 Reductions in of mental illnesses and suicidal ideation - Reductions in mental health-related sequelae Reductions in re-attempts Increases in connectedness II Improvements in coping skills I Increases in identification of individuals at?risk for suicidal behavior I Increases in at?risk individuals in treatment I Increases in community members trained to identify at-risk individuals - Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. 32 Gatekeeper training. Applied Suicide Intervention Skills Training (ASLST) is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, Kleinman (2013] evaluated the training across the National Suicide Prevention Lifeline network of hotlines over the period 2008- 2009. Using data from 1,410 suicidal individuals who called 17 Lifeline centers, the researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors without training in were significantly more likely to feel depressed, suicidal, more overwhelmed, and less hopeful by the end of their call to the hotline compared to those with training in Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASLST did not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013). Gatekeeper training has also been a primary component of the Garret Lee Smith (GL5) Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GLS trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GLS trainings had significantly lower youth suicide rates one year following the training implementation {Walrath, Garraza, Reid, Goldston, 81 McKeon, 2015}. This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GLS program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GLS program than did similar counties that did not implement GLS activities (4.9 fewer attempts per 1000 youths; Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015). More than 79,000 suicide attempts may have been prevented during the period examined, following implementation of the GL5 program. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Nationai Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in 33 pain, hopelessness, and intent to die between initiation of the call [time 1) to follow-up [time 3). Between time 2 {end of the call) to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die [Gould Kalafat, Harrismunfakh, 8: Treatment for people at-risk of suicide. The improving Mood Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow?up (biweekly during an acute phase and during continuation phase) by a depression care manager (Hunkeler et al., 2005). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24?months of follow-up [Hunkeler et al., 2006: Unutzer et at, 2005] relative to patients who received care as usual. Collaborative Assessment and Management ofSulcidallty (CAMS), is a therapeutic approach for suicide?specific assessment and treatment. The program's flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (Jobes, 2012), in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual {intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow?up [Comtois et at, 2011). Other examples include Dialectical Behavioral Therapy and Attachment-Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year follow?up than women receiving community treatment [23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2005]. 34 ,{Comment Formatting is weird hereKleinman, 2007i, a program for adolescents aged 12-13 and is designed to treat clinically.I diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxietv {Diamond et al., 2010]. A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly,r greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally. a significantlv higher percentage ofABi-T participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care [69.2% vs. 34.6%} and at 24 weeks (82.1% vs. 46.2%] (Diamond at al., 2010]. The Veterans Affairs Transioting initiatives for Depression into E?ective Saiutians project (TIDES) uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primaryr care setting, where most patients are first detected and subsequentlv treated for manv mental health conditions. An evaluationthu-evgll of prefee?speeifiealivrfound signi?cant decreases in depression severitv scores among 20% of primarv care patients [Rubenstein et at, 2010). TIDES also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively (Rubenstein et al., 2010}. Treatment to prevent re-attempts. Several strategies that aim to prevent re?attempts have I demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief, intervention with Foiiow?up Visits is a program that involves a one~hour discharge information session that addresses suicidal ideation and distress, risk and protective}I factors, alternatives to self?harm, and referral options, over 18 months (at 1, 2, 4, 11 weeks and 4, 6, 12, 13 months]. Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 13-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, lran, and China] found that a brief intervention combined with Si follow-up visits over 18-months was associated with significantlv fewer deaths from suicide relative to a treatment?as?usual group versus respectively) (Fleischmann et al., 2003]. Another example of treatment to prevent re-attempts involves active faiiow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically leg, or every few months in the first 12 months post- discharge with some programs continuing contact for two or more vears}. In a meta-analysis 35 {Comment IAI: Attempts? Drshould this sav "self-harm?? conducted by Inagaki et al. {2015], interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts (Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CRT-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). 36 Comment Need to modify the rationale so that it speaks to the primary Lessen Harms and Prevent Future Risk [Rationalci fa? approaches included below - perhaps starting with the second sentence and _expanding on that a bit. . . . . Comment w: This introductory statement needs to be modified now that this section only includes postvention and ksafe reporting following a suicideare suicide every year in the United States and throughout the World. Risk of suicide and suicide risk factors shown to increase among shesegeople who have lost a friend!p_e_er, family member, co-worker, or other contact to suicide (Pitman, Osborn, King, St Erlangsen, 2014). intentions, mMedia and others responding to suicide may inadvestensly-add to this risk-amongshe reporting regardiegon suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to @13th knongs_suicide contagion (Etzersdorfer 8: Sonneck, 1993; Niederkrotenthaler 8L Sonneck, the evidence is still being byilt in this area, particLLliirly with regard to the impact of policy and practices on suicide and suicide attempts in the United States. Iakingeffeetivemeasures to care for thisthe bereaved population through such means as postvention interventions leg. counseling, support groups and debriefing sessions} and safe reporting on suicide shown impacts in other countries. Approaches approaches that can be used to lesson harms and reduce future risk of suicide includgi-ng WWmemm?m?a?ng; for the bereaved; and safe reporting following a suicideiaeseaved,_ . For example, research suggests that eExposure to sensationalized or otherwise uninformed Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief (Szumilas 3-: Kutcher, 2011). It Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline}, and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide 3? contagion. Fesponsilsle and- safe Fepeet+ng mavv se-leiele 1 .f 1! Comment IAI: Suggest ?ashing this out a bit so that the reader has a good sense of what is meant bv ?safe reporting". Comment IAI: Outcome section needs 4? to be modi?ed now that only postvention and safe reporting are included. modi?ed Potential butcomeg Reductions in suicidalit?ideation/attempts} Reductions in distress Increases in treatment seeking Improvements in messaging?reporting following suicide Reduetmns-Mrue-at-tempts Reductions in contagion effects related to suicide Evidence Eurrent evidence suggests that lessening harm tllgoughposgve?tiola?d safe reportingcan impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, StondBv Response Service l5tondBy}, provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing communitv services matched to their needs (Visser, Comans, 8: Scuffham, 2014}. In a study by Visser et al. [2014}, StondBy clients were significantly less Iikelv to be at high risk for suicidalitv (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the Stond?v program {48% and 64% respectively]. Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services] (Cerel 8: Campbell, 2003}. Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging I about suicide is to encourage news media adhere to Recommendations for Reporting on Suicide {Comment IAI: Statement needs to be I Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline], and that avoid sensationalizing events or reducing suicide to one cause, 33 can help reduce the likelihood of suicide contagion. The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually {Niederkrotenthaler 8r Sonneck, 2007}. Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010). Reports of individual suicidal ideation not accompanied by reports of suicide or suicide attempts, along with reports describing a ?mastery? of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports (Niederkrotenthaler et al., 2010). Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, mghealth sector (including insurersL-and?providers, and health systems], and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address 39 some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non- governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront ofmany community?based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Probiem-Soiving Skiils to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support Peopie fit-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 40 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andfor medicolegal standards (AE. Crosby, Ortega, et al., 2011). uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems A.E. Crosby, Ortega, et al., 2011). Surveillance systems exist at the federal, state, and local levels. it is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System (Ni/55) and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of?death data from death certificates. is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016}. Data from state and local Child Death Review teams and Suicide Death Review Teams {which are in a few states] offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic injury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department). in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade 41 students and is a key resource in monitoring health?risk behaviors among youth, including whether youth Comment IAI: Last sentence added per suggestion of EDIE. You can cite D?v?P's new have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by i Strategic Vision. a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2313}. The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Heoith is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state?level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs}; mental health [past year mental illness, co?occurring illnesses]; and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk. It is also important at all levels (local, state, and federal] to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long? term outcomes are an important part of program evaluation. The evidencevbase for suicide prevention has advanced greatiy over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum], as opposed to merely examining their effectiveness on risk factors. More research is also needed to examine the effectiveness of primary prevention strategies {before risk occurs} and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic efFECts within a comprehensive prevention approach. w, there are also many potential opportunities to build and strengthen partnerships across program areas leg, violence prevention, substance abuse prevention] to evaluate the impact of different approaches on multiple butcomedConclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis - at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help?seeking, 42 mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 43 References Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., . . . Coventry, P. (2012). Collaborative care for depression and anxiety problems. The Cochrane Library. ArsenauIt-Lapierre, G., Kim, (2., 8L Turecki, G. (2004). diagnoses in 3225 suicides: a meta-analysis. BMC 4, K. E., Canham-Chervak, M., Black, S. A., Dada, E. 0., Millikan, A. M., S: Jones, E. H. (2012). Mental health risk factors for suicides in the US Army, 2007?8. injury Prevention, 18(6), 405-412. Ba ndura, A. (1986). Saciaifoundatians of thought and action: A social cognitive theory: Prentice-Hall, Inc. Barker, E., K?lves, K., 8i. De Leo, D. (2014). Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research, 18(3), 227-240. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi:10.1080f00048620902823714 Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukol, B. (2014). Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the Wario' Health Organization, 92(9), 641-655. Beyer, K. M., Layde, P. M., Harnberger, L. K., 8; Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49-64. doi:10.1127f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 3: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i?-ill. Bossarte, R. M., Karras, E., Lu, N., Tu, R., Stephens, B., Draper, J., 8: Kemp, J. E. (2014). Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Heaith Rep, 129(6), 516?525. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. El Ten Have, T. R. (2011]. A difference? in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol, 1 74(11), 1296-1306. doi:10.1093lajeikwr273 Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. R., Hollander, J. E., Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?5?0. Bruce, M. L, Ten Have, T. R., Reynolds C. F., Katz, 1. Schulberg, H. C., Mulsant, B. H., . . .Alexopoulos, G. S. (2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291(9), 1081?1091. Capalcli, D. M., Knoble, N. B., Shortt, J. W., Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., E: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full version?a.pdf Centers for Disease Control and Prevention. (2016). Web?Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: 44 Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. Aicohoi: Science, Poiicy and Pubiic Heaith, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 5513), 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directed Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. {2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and iife-threatening behavior, 32(3), 321-328. Cunningham, P. i. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff 28(3), w490-501. Curtin, S. C., Warner, M., St Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 45 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 81'. Levy, S. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Doran, C. M., Ling, R., Gullestrup, J., Swannell, 5., 8: Milner, A. (2016). The Impact ofa Suicide Prevention Strategy on Reducing the Economic Cost of Suicide in the New South Wales Construction Industry. Crisis, 37(2), 121-129. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., Giles, W. H. (2001). Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA, 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1937}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8t G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 1897). Edwards, R. W., JumpervThurman, P., Plested, B. A., Getting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 2918,02. Escobeclo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 689694. Etzersdorfer, E., 81 Sonneck, G. (1993). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Felitti, V.J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks,J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245-258. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., 8.: Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi:50042? 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., 8t Zhou, C. (2015). Estimated lifetime medical and workeloss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 6488), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 8: Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005?2010. Am Public Health, 105(2), 311-316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi:10.2105lAJPH2013301608 Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 46 Gilbodv, 5., Bower, P., Fletcher, J., Richards, 0., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternai medicine, 166(21), 2314-2321. Godoy Garraza, L., Walrath, C., Goldston, D. 6., Reid, H., St McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. (2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. doi:10.1111isltb.12049 Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. Grossman, D. C., Mueller, 6. A., Riedv, C., Dowd, M. 0., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Haegerich, T. M., 8: Dahlberg, L. L. (2011). Violence as a public health risk. American Jaurnai aftifestyie Medicine, 1559822611409127. Hailey, D., Roine, R., at Ohinmaa, A. (2008). The effectiveness of telemental health applications: a review. The Canadian Joarnai of 53(11), 769278. Hambv, 5., J. (2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse Briefs in Sacioiogy. New York, NY: Springer. Han, B., Crosbv, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., 8; Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 126-186. Harris, E. C., at Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205228. Harris, E. C., 8: Barraclough, B. (1993). Excess mortality of mental disorder. Bri 173, 11?53. Harris, K. M., Carpenter, C., 3: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8; Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital~treated selfnpoisoning. British ioarnai of 198(4), 309-316. Hawkins, J. 0., Catalano, R. F., 8-1 Kuklinski, M. R. (2014). Communities that ca re Encyciopedia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Haves, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 75(4), 431-456. Haves, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int;r Law 138-194. doi:10.1016ij.ijlp.2013.04.010 Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring lnternalizing Q, 26(3), 189-201. Houle, J. N., St Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am i Pubiic Heaith, 104(6), 1073-1079. 47 Hunkeler, E. M., Katon, W., Tang, L., Williams, J. Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(7536), 259-263. Inagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad201412.048 lobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8.: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12], 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. (2015). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 1-174. doi:10.15585immwr.ssE550631 Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. S., Wang, W., Toyinbo, F., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55?528. Kessler, R. (2., Demler, 0., Frank, R. (5., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515?2523. doi:10.1056lN Elk/153043266 Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 223?279. doi:10.1027l0227? 591013000136 Klomek, A. E., Sourander, A., 8; Gould, NI. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 13?6. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), ?3u88. Leeb, R. T., Lewis, T., 81 Zolotor, A. J. (2011). A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of [lifestyle Medicine, 454-468. Lineberry, T. W., O'Connor, 5. S. Suicide in the US Army. Mayo Clinic Proceedings, 871-378. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 757-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. 48 Maimon, D., Browning, C. 8., 8; Brooks-Gunn, i. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Sherri, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8c Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. {2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention :8 Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. G., 8: Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8: Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243-253. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21? 41. Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 leson, M., Gerhard, T., Huang, C., S., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Perron, S., Burrows, 5., Fournier, M., Perron, P. A., Duellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 Pirkis, 1., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 8: Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. 49 Pitman, A., Osborn, 0., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Plested, B. A., Edwards, R. W., 31. Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rittenhouse, D. R., Fryer, G. E., Phillips, R. L., Miyoshi, T., Nielsen, C., Goodman, D. C., Grumbach, K. (2008). Impact of Title VII training programs on community health center staffing and National Health Service Corps participation. The Annais of Famiiy Medicine, 392-405. Rowhani-Rahba r, A, Simonetti, J. A, 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiai Rev, 38(1), 111-124. Rubenstein, L. V., Chaney, E. F., Ober, 5., Felker, B., Sherman, S. E., Lanto, A., at Vivell, S. (2010}. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Systems, Heaith, 28(2), 91. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emeryr Med, 17(1), 8-14. Russell, 5. T., Joyner, K. {2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Jaurnai of Pubiic Heaith, 91(8), 1276-1281. Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, G. 84 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352?362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and iife? threatening behavior, 25(1), 10?2 1. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 4959. Spoth, R. L., Guyll, M., Day, 5. X. (2002). Universal family+focused interventions in alcoholsuse disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 228. Stack, 5., 8t Wasserman, l. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and We threatening behavior, 37(1), 103-112. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. FL, 8: Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. .i Correct Heaith Care, 21(3), 222u242. Sudak, H., Maxim, K., 8! Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. doi:10.1176fappi.ap.32.2.136 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Pubiic Heaith, 102(1), 18-29. Toosi, M. (2015). Labor force projections to 2024: The iaborforce is growing, but siowiy. Washington, Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, G. M., 8: Crawford, M. J. {2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2016. Retrieved from 50 U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Clinician Retention: A Story of Dedication and Commitment. Retrieved from U.S. Office of the Surgeon General, at National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, DC. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. 5., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journaiof Community 42(1), 19-28. Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2015}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(sszz), Webster?Stratton, C., Hammond, M. Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consuit Ciin 65(1), 93?109. Webster?Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283302. Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible veers teacher and child training programs in highsrisk schools. Journal of child and 49(5), 471?488. Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., St. Anthony, J. C. (2008). The impact of two universal randomized first? and classroom interventions on voung adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?23. Wilkins, N., Tsao, B., Hertz, M., Davis, H., 8L Klevens, J. (2014). Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. World Health Organization. (2013). Giobal and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and Non?Partner Sexual Violence. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurrav, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8; Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment benefit programs v? 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization v, state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity Laws 1/ 1/ Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps (NHSC) suicide care Healthcare Telemental health (TMH) Social services Safer sunclde care through systems change Henry Ford Perfect Depression Care {Pre- v, V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt intervening at suicide hot spots Government (local, state) Create - Safe storage practices . protective Public Health environments Emergency Department Counseling on v, Organizational policies and culture 50 Best Available Evidence Together for Life US Air Force Suicide Prevention Program Correctional suicide prevention Business/Labor Justice Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Alcohol outlet density 1/ Government (local, state) Business/labor Promote connectedness Peer norm programs Public Health Sources ofStrength Education Communit en a ement activities 5 Public Health Greening vacant urban spaces Government {local Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mental Health Program Education Good Behavior Game Parenting skill and family relationship approaches Public Health The incredible Years Education 51 Strengthening Families 10-14 Best Available Evidence Identify and at-risk support people Gatekeeper training Public Health Applied Suicide intervention Skilis Training Healthcare Garret Lee Smith Federal Grant Program v? Crisis Intervention Public Health Soc'al Services National Suicide Prevention Lifeline v? I Treatment for people at risk of suicide improving Mood - Promoting Access to 1/ Collaborative Treatment Collaborative Assessment and Management Healthca re of Suicidality (CAMS) Dialectical Behavioral Therapy (DST) Somal Servlces Justice Attachment-Based Family Therapy v? Translating initiatives for Depression into Ejj?ective Solutions project Treatment to prevent re-attempts ED Brief intervention with Follow-up Visits Health ca re Active follow-up contact approaches Social services CBTfar Suicide Prevention 52 Best Available Evidence Postvention Lessen harms StondBv Response Service 1/ 1/ ea care and prevent future risk Safe reporting following a suicide Public Health 1/ Media Gurdelmes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgments We would like to thank the following individuals who contributed in specific ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely,r not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this FESOUFCE. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access to mental health care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and intervening to lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, Melanson, 2011). Self- directed violence may be suicide! or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the US (Centers for Disease Control and Prevention, 2016). Overall suicide rates increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2014, the rates for these groups were 17.8 and 16.4 per 100,000 population, respectively (Centers for Disease Control and Prevention, 2016). Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 48% from 1999 to 2014, with steep increases seen among both males and females aged 45-64 years; (Curtin et al., 2016); Veterans and other military personnel, whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in decades et al., 2012; Lineberry 8: O'Connor); workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; McIntosh et al., 2016); and lesbian, gay, bisexual, and/or queer youth, who experience increased suicidal ideation and behavior compared to their heterosexual counterparts (Kann et al., 2016; Russell Joyner, 2001). Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8i. Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: History of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: High conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: Availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (US. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. it is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office ofthe Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (U.S. Office ofthe Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) increases the risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts (Bossarte et al., 2014; D. P. Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, 8: Gould, 2010; Leeb, Lewis, 8: Zolotor, 2011; World Health Organization, 2013). Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, increases the risk for suicide and suicide attempts several fold (Bellis et al., 2014; Dube et al., 8 2001). The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide (Haegerich 8c Dahlberg, 2011; Hamby 8.: 2013; Wilkins, Tsao, Hertz, Davis, at Klevens, 2014). Further, just as risk factors may be shared across suicide and violence, so too may protective factors overlap. For example, connectedness to one?s community (Kleiman, Riskind, Schaefer, Weingarden, 2012), school (Carter, McGee, Taylor, St Williams, 2007), family (Maimon, Browning, Brooks-Gunn, 2010), caring adults (Capaldi, Knoble, Shortt, 8L Kim, 2012; Losel Farrington, 2012), and pro-social peers (Wyman et al., 2010) enhances resilience to suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). In an early study, Crosby and Sacks (2002) estimated that 7% of the U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. in a more recent study, in one state, Cerel et al (2016) found that 48% ofthe population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (A. L. Chapman Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, 8L Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, Carpenter, 2008). Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegerich, Luo, 8: Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alone (Florence et al., 2015}. Adjusting for potential under-reporting ofsuicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, Silverman, 2016). The overwhelming burden of these costs were from lost productivity over the life course, with the 9 average cost per suicide being over $1.3 million {Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (US. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman E: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public le.g., business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014L Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available 10 IAI: Afew nth??hlnisfha! evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, cauid'bemen?oned: and even those that are effective may not work across all populations. Tailoring programs and more . evaluation may be necessary to address different population groups. The evidence?based programs, :5 practices, or policies included in the package are not intended to be a comprehensive list for each i paragraph; approach, but rather to serve as examples that have been shown to impact suicide or have beneficial {if "WWHandappwachesdeimeate effects on risk or protective factors for suicide. i: prevention 6:1er If?? immt-varfom?v 3 reared surnames. Marrow-arena: Context and Cross?Cutting [Theme If ?g?gf??mfm?gmf One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches that have been included in this technical nmjguymijme?a??mafwg package represent different levels of the social ecology, with efforts intended to impact the community strategy eta-change schao?. and societal levels, as well individual and relationship levels. The strategies and approaches are intended strategy sometimes have components to work in combination and reinforce each other to prevent suicide [see box below). The strategies are thatmather strategies. For example, arranged in order such that those strategies hypothesized to have the greatest potential for broad public . . . Mob?izing Men-and desms?i?egoh health Impact on solclde are Included first, followed by those that might Impact more select populations ?m persons who have already made a suicide attempt]. thatPretEct?mimt Wisteria: strategy. inaludes?zsterfng healthy dating - - - Maharishi? whicalrs nisa?mnd in some "?ammg su'c'de the Teach saris APPma?h to Fianna Waterman." Strategy Strengthen economic supports I Strengthen household financial security I Housing stabilization policies Coverage of mental health conditions in health Strengthen access and delivery of suicide insurance policies care I Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments 0 Reduce access to lethal means among persons at- rislc of suicide Organizational policies and culture Com mu nity-based policies to reduce excessive alcohol use I Peer norm approaches Community engagement activities Socialaemotional learning programs Parenting skill and family relationship approaches Promote connectedness I Teach coping and problem-solving skills a I I Gatekeeper training Crisis Intervention Screening and treatment for people at risk Treatment to prevent re?attempts Postvention Safe reporting following a suicide Identify and support people at risk Lessen harms and prevent future risk 11 It is important to note that these strategies are not mutually exclusive. Thus, examples of programs, policies, or practices listed under one strategy may also be relevant to another strategy. For instance, some forms of crisis intervention, an approach under identify and Support People ?it-Risk, may also be considered as ways to Lessen Horms and Prevent Future Risk. Social emotional learning programs, an approach under the Teach Coping and Problem-Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced [Turecki, 2014}. Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Getting, Swanson, 2000; Hawkins, Catalano, 3; Kuklinski, 2014; Plested, Edwards, 8: Jumper-Thurman, 2006). These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program 12 materials, training and technical assistance] can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program le.g., workplace policies; screening combined with care management). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems; buffering these risks can therefore, potentially protect against suicide (Stack 8: Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes Reductions in foreclosure rates 14 Reductions in eviction rates Reductions in emotional distress Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federai?Stote Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss (Cylus, Glymour, 8c Avendano, 2014). An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: $7,990 per person in US. constant dollars; Cylus et al., 2014). The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment {less than 5 weeks, 5?14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk (Classen 8a Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance] have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families - TAN F) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. In terms of lives saved, Flavin 3t Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year {Flavin El Radcliff, 2009). Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable 15 individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The National Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslyr evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014). Another studv of data from 16 US. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began) to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery of Suicide Care Rationale While most people with mental health problems do not attempt or die by suicide [Olfson, Gerhard, Huang, 8: Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {ArsenauIt-Lapierre, Kim, 81 Turecki, 2004; E. C. Harris 81. Barraclough, 1997; Tyrer, Reed, 8L Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 8: Barraclough, 1998; World Health Organization, 2014). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shern, Bagalman, 8t. Cao, 2007}. Findings from the National Comorbidity Survey indicate that relatively few people in the US. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to suieide prevention {World Health Organization, 2014}. Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care (CE. Coffey, 2007). Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches One approach to strengthening access to mental health care is through the provision of mental health coverage in health insurance policies. Attending to training and provider shortages is another approach. - Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on parwith coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co?pays, deductibles, inpatientfoutpatient services, prescription drugs, and hGSpitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. if a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. I Reduce provider shortages in underserved. areas. Access to effective and state?of~the?art mental health care is largely dependent upon the training and the size of the mental health care Over 85 million Americans live in areas with an 17 insufficient number of mental health providers; this shortage is particularly severe among low- income urban and rural communities (U.S. Department of Health and Human Services Health Resources and Services Administrations, 2016a). There are a number of ways. to increase the number and distribution of practicing mental health providers including providing and expanding existing state and federal programs; increasing wages and reimbursement ratios for mental health services; and expanding telemental health services. Such approaches can increase the likelihood that those in need will be able-to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide however this is just one piece of the puzzle. Care being accessed must also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments [see identify and Support People At-Rr?sk. p.311, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance (National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014), Potential Outcomes I Increases in access to mental health services 0 Increase in utilization of mental health services 0 Reductions in of mental illnesses and suicidality - Reductions in rates of suicide attempts I Reductions in rates ofsuicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006} found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013) examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated 18 19 laws, equated to the prevention of 592 suicides per year (Lang, 2013]. with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity {Comment IAI: May need to rework the treatment attrition rates were lower among patients receiving telephone- introductory statements. administered patients receiving facade?face therapy; may not only offer improved access to mental' healthoare, but, it'rna?y also ensure-cont] unity of care, and thereby further reduce the risk for suicidE. '11? i I i? 1? Safer suicide care through systems change. {Studies have demonstrated thatmllaborative care medals are as effective in treating depression and suicidal ideation as stagdarcl- care {Archer et ,5 at, 2011,- Bruce et .11., 2004,- G?bady, ?ower, Fletcher. ?lthards. En Simian, 2005i, . Henry Ford health ca re system, which is a large health maintenance organization (HMO) in the state of Michigan pioneered the Perfect Depression Core program,_the pre-cursor to what is now called Zero Suicide, T_he_overa l goal of Peg?ect Depression Core was to eliminate suicide among 3M9 members. More broadly. the goal ofthe program was to redesign delivery of depression care to achieve "breakthrough improvement" in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, ef?ciency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow?up care system wide (C. E. Coffey, 2005]. An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% E. Coffey, 2006; C. E. Coffey, Coffey, 3: Ahmedani, 2013}. Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1990 to 2010 {110.3 to 42.6 per 100,000; p<.04} with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 ip<.001] (M. Coffey, Coffey, 8t. Ahmedani, 2015}. 2i] Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes (Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide (Dahlberg 8: Krug, 2002; US. Office of the Surgeon General 81'. National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults (who comprise 42.6% of the workforce; Toosi, 2015} andasealso elevateda'eramong certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016], and among people in detention facilities leg. iail, prison). to name a m. Thus, woskplaeesrsettings where these populations work and residemay seweesaa? ideal setting for reaching programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015}. Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition (Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; A.E. Crosby, Espitia-Hardeman, Ortega, 8: Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8: Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Eyck, &Teplin, 2015). Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. I Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 35% of people who use a firearm in a suicide attempt will die from the injury}. Research also indicates that 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001) and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms-docked in a secure place leg, in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8t Rivara, 2016; Runyan et al., 2015). Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments Leg. residentialprograms). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, eha-nging?positive social norms, assessment referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention, and safe physical environments. policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides (Escobedo Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges, Wilcox, 2004). Potential Outcomes Increases in safe storage of lethal means Reductions in suicide attempts Reductions in suicide deaths 22 Increases in help-seeking Reductions in alcohol-related suicide deaths Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015}. For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, 31 Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement ofsuicides to otherjumping sites {Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removoi of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold (Beautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani?Rahbar et al., 2016). Another program, The Emergency Department Counseiing on Access to Lethoi Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2016} found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 8: Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 100,000; Mishara 8; Martin, 2012}. Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement}. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (Knox, Litts, Talcott, Feig, 8r Caine, 2003]. Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (Knox et al., 2003}.The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (Knox et al., 2003). A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch) found significantly lower rates of suicide after the program was launched than before (Knox et al., 2010). These effects were sustained overtime, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (Knox et al., 2010}. Mates in Construction (MIC) is a multifaceted suicide prevention strategy developed in Australia and implemented and evaluated in construction workglaces. The program has three primary components: general awareness training, connector training, and applied suicide intervention skills training (ASIST). A one-hour general awareness training is provided for all construction workers on site with the goal of increasing awareness of suicide as a safety issue and a problem facing workplace health, increasing knowledge about warning signs of suicide, and encouraging help-seeking behaviors among construction workers. A four-hour connector training session is provided for individuals labeled "connectors?, who serve to keep coworkers safe while 24 connecting them to help through an ASIST-trained worker, MIC field manager,_or case manager. Individuals trained in ASIST take part in an intensive two-day training to build skills for identifying and responding to suicidal ideation and behavior, with the goal of implementing a safety plan to involve at-risk individuals obtaining the care they need in a safe environment. An evaluation examining the potential economic impact of widespread implementation of MIC among New South Wales construction industry workers identified suicide rates before MIC implementation from 2008-2012 and after implementation of MICfrom 2013-2017 usingtime series data on NEW suicide deaths and the construction workforce. Doran et al. [2016) found that MIC could potentially avert 0.4 suicides, 1.01 full incapacity cases, and almost 5 short absences, per year, in the construction industry, generating an annual savings of $3.66 million AU. They further indicated that ever?ustralian dollar invested in MIC would result in a $4.60 return {Doran, Ling, Gullestrup, Swannell, 8: Milner, 2016}, Finally, whileWh?i?le the evidence is still beingbuilt for suicide prevention in correctional facilities, preliminary evidence exists for comprehensivepolicies and practices within these settings. These policies and practices include: routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols in place, notification of suicidal behavior/suicide through the chain of command, and critical incident stress debriefing and death review. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates dropped from a rate of before the intervention to the following year (Hayes, 1995). Other similar programs have seen declines in suicide both in the United States and internationally (Barker, K?lves, De Leo, 2014). Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, speci?cally, is related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, Remer, 2009}. 25 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951}. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 8; Krug, 2002}, for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capitol refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, 8L Zhu, 2013]. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest the pattern is towards a positive association between social capital measured by social trust, community/neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole {Centers for Disease Control and Prevention, 2009). Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer norm approaches seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 26 enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors - Increases in referrals for youth in distressed Increases help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm approaches. Evaluations show that programs such as Sources of Strength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources ofStrength conducted with 18 high-schools (E metropolitan, 12 rural), Wyman et al. (2010) found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010}. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism {Branas et al., 2011). 27 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014). Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness) characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004). Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on developing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use) associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, 8L Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children?s behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual or groups of families. Some programs have sessions primarily with parents while others include sessions for parents, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes ofchild development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes Comment Should we add references for some of the svstematic reviews on school and parentingj'famllv programs? Comment IAI: Dropped Signs of Suicide Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors li.e., depression, ansietv, conduct problems, substance abuse] I i Consider adding Two examples - mav be ?ne here, so not sure we need to Improvements in help-seeking behavior add another program. Improvements in social competence and emotional regulation skills Improvements in problem?solving and conflict management skills . i i rigorous evaluations to Improve resilience and reduce risk factors for various behavior including ones Evidence Several social emotional learning and parenting and family relationship programs have been shown in closelv related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. KnoxBurkhart, Hunter, 2010). a Social emotional learning rogramd. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers to teach adolescents about the risk and protective factors associated with suicide {including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wasserman et al., 2014). In a cluster-randomized controlled trial conducted across 10 European Union countries and 158 schools, students in schools randomized to YAM were significantlv less likelv to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which receiVed educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was red Liced bv over 50% demonstrating that out of 1000 students, five attempted suicide in the MM group compared to 11 in the control group. Additionallv, related to severe suicide ideation, in the MM group absolute risk fell bv 0.50% and relative risk fell bv 49.6% {Wasserman et al., 2014). Another example is the Good Behavior Game (GEE), which is a classroom-based program for elementarv school children aged 6?10. The program uses a team?based behavior management strategy that promotes good behavior bv setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 656 program is to create an integrated classroom social svstem that is supportive of all children being able to learn with little aggressive or disruptive behavior [Wilcox et al., 2008}. Two cohorts of vouths participated in the program in 1935-86 and 1986-8? school vears when thev were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 29 (336 when they were in the first grade reported half the adjusted odds of suicidal ideation and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of 656 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions (Wilcox et al., 2008). The researchers believed this may have been due to a lack of implementation fidelity. 636 was also found to be associated with reduced risk of later substance abuse, 3 risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self-regulation and social competence {all protective factors for suicide) (Herman et al., 2011). The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. liliebster-Stratton, Reid, 81 Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8: Hammond, 1997; C. Webster-Stratton, Reid, Hammond, 2001). The program demonstrated greater benefits as the dosage ofthe intervention increased (Herman et al., 2011). Additionally, Strengthening Families 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guyll, 8: Day, 2002). Strengthening has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 30 Comment IAI: Need to tweak rationale so that itincludes the treatment pieces. Identify and Support People AtsRisk J, Comment IAI: There was a comment that people who are disadvantaged sounded awkward which I agree. What i i i about marginalized? i: Comment We?re no longer giving .5 [Rationale] In order to decrease suicide, attention to people at increased risk is necessary, as these individuals tend to experience suicidal behavior at higher than average rates. These vulnerable or disadvantaged populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have attempted suicide previously; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are ch examples of screening {that went with Henry Ford] so I took that out. homeless; individuals of sexual minority status; and members of certain racial and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; Curtin et al., 2016; Kann et al., 2015; Lineberry 8t O'Connor; Russell 8: Joyner, 2001}. Supporting these vulnerable groups requires proactive case finding along with access to, and retention in, mental health services. Finding effective ways of identifying at-risk or vulnerable groups, customizing services to make them accessible and engaged in care remain key challenges. For example, simply improving services does not guarantee that those services will be used by those most in need of them, nor will it necessarily increase the number of people who follow treatments that are recommended. I l' I bopulation? may face social and economic issues that mavcan adversely affect their ability to access categorise-statem- Approaches The following three approaches focus on identifying and supporting people at increased risk. i i I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, 3 primary and urgent care providers, and others in the community to identify people who may be i at risk of suicide and to respond effectively, including facilitating treatment seeking and support i services. Gatekeeper training may be impiemented in a variety of settings to identify and support i i i people at risk. - Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at-risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or inrperson. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Like means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Screening and treatmentTreatment for people at?risk of Euicidd; Sereening canine- usedin; as suieide an} identified and? appsepnate ease. This can include various forms of 31 Potential Outcomes Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of support for these individuals can positively impact suicide and its associated risk factors. I I I delivered by licensed providers to help individuals with mental health problems and other suicide risk factors taesuieide?with problem-solving and napolsiv-ity?and?emotion I I I regulation. Treatment usually takes place in a one on one or group format between patients and cliniciani easesas needed. Treatment that employs collaborative lie, betwee_n patient and therapist o_r Comment IAI: Wonderif it wouldjust and_can vary in duration from several weeks to ongoing therapy, foeyeaesinseme be better to end the sentence after the word ?clinicians"? EH: Yes; I'thinlt there fine; Comment Moved ASIST here per reviewers comments. JL Dropped MHFA consider incorporating care manager] and/or integrated care leg. linkage between primary care and behavioral health carel can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk et al., 2012; Bruce et al., 2004; Gilbody, Bower, Fletcher, Richards, 8: Sutton, 2006]. Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities leg, home visits, mail, telephone, e?mail] to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on?one interpersonal therapy andlor group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296-2596 reattempt within a year, and 356-998 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (lnagaki et al., 2015] Reductions in suicide attempts Reductions in suicide deaths Increases in identification of individuals at?risk for suicidal behavior Increases in at-risk individuals in treatment Increases in community members trained to identify atvrisk individuals Increases in referrals for health care Gatekeeper training. Applied Suicide Intervention Skills Training is a widely implemented} training program that helps hotiine counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a 32 Lthat program in another category? randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman (2013] evaluated the training across the Notionoi Suicide Prevention Lifeiine network of hotlines over the period 2008- 2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, the researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors without training in were significantly more likely to feel depressed, suicidal, more overwhelmed, and less hopeful by the end of their call to the hotline compared to those with training in Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in did not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013). Gatekeeper training has also been a primary component of the Garret Lee Smith Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GLS trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation (Walrath, Garraza, Reid, Goldston, McKeon, 2015). This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GLS program activities also had significantly lower suicide attempt rates among youth ages 16 to 23in the year following implementation of the GL5 program than did similar counties that did not implement GLS activities (4.9 fewer attempts per 1000 youths; Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015). More than 79,000 suicide attempts may have been prevented during the period examined, following implementation of the GLS program. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days lmean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1] to follow-up (time 3). Between time 2 {end of the call) to time 3, the effect remained for pain and hopelessness, but was not significant for intent to die (Gould, Kalafat, Harrismunfakh, 8t 33 Kleinman, 2007). improving Mood Promoting Access to Collaborative Treatment to prevent suicide among oider primarv care patients bv reducing suicide ideation and facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-Up (biWeeklv during an acute phase and during continuation phase} bv a depression care manager {Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up {Hunkeler et al., 2006; Unutzer et al., 2006?: relative to patients who received care as usual. Assessment and Management ofSuicidoiity (CAMS), whieheis a therapeutic approach for suicide-specific assessment and Th_e program?s flexible approach can be Used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. ?essions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 5 correlational studies (lobes, 2012), in a varietv of inpatient and outpatient settings and in one RCT with several additional RCTs under way. CAMS has been associated with significant improvements in suicidal ideation, overall distress, and feeiings of hopelessness at 12 month follow-up among a community-based sample of suicidal outpatients. lComtois et al., 2011). Other examples include Dialectical Behavioral Therapy (DST) and Attachment-Based Family Therapy a multicomponent therapl,r for individuals at high risk for suicide and who may struggle with impulsivitv and emotional regulation. The components of BET include individual therapv, group skills training, between?session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self~ injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year followvup than women receiving communitv treatment [23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined [Linehan et al., 2006}. a program for adolescents aged 1243 and is designed to treat clinicallv diagnosed maior depressive disorder, eliminate suicidal ideation, and reduce dispositional anxietv {Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT 34 1 Screening and It-reatment for people at?risk of suicide. esampie et- a- treatmene with r" - Comment IAI: May need to rework introductorv statement and description to fit in the notion of screening. experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%} and at 24 weeks {82.1% vs. 46.2%) (Diamond et al., 2010). the Veterans Affairs Transiao?ng initiatives for Depression into E??e?ctive Solutions project (VA a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. Mllaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are ?rst detected and subsequently treated for many mental health conditions. A study of the VA TIDES project, specifically, found significant decreases in depression severity scores among 70% of primary care patients. also demonstrated 85% and 95% compliance with medication and follow-up visits, respectiVely (Ruhens?tein et al., 2010}. Treatment to prevent re-attempts. Several strategies that aim to prevent re-attempts have 1.. DqulLi and referral options, combined with nine follow- up contacts over 18 months (at 1,2,4, 7,11 weeks and 4, 6, 12,18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18-months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency dapartrnents in five countries (Brazil, India, Sri Lanka, lran, and China) found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively) {Fleischmann et al., 2008}. Another example of treatment to prevent re-attempts involves active follow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically g. or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years]. In a meta- -ana ysis conducted by lnagaki et al. (2015), interventions to prevent repeat asttempt in patients admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 12% for up to 12 months post-discharge; however, the 35 effects of these approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of the post-crisis suicide prevention long-term follow-up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts {Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016L Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-5P) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties} leading up to and following the suicidal event; safety plan development; skill building; and also has family skill modules focused on family support and communication patterns as well as improving the family?s problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). 36 Lessen Harms and Prevent Future Risk {Rationale} Individuals who have experienced mental health challenges, suicidal ideation, who have made suicide attempts or engaged in non-suicidal self-injury are at increased risk of suicide (US. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012]. Risk of suicide can also increase among those who have lost a friend, family member, co-worker, or other acquaintance to suicide lPitman, Osborn, King, 31 Erlangsen, 2014). Exposure to sensationalized or uninformed reporting regarding on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion [Etzersdorfer 8: Sonneck, 1998; Niederkrotenthaler 8: Sonneck, 2013?]. Approaches including various therapeutic treatments and approac?1 here are a number of approaches that can be used to lesson harms and reduce future risk of suicide es providing continuity of care, caring for the bereaved, and safe reporting following a suicide. 'x Comment IAI: Need to modify the rationale so that it speaks to the primary approaches included below - perhaps starting with the second sentence and expanding on that a bit. iEEC-romant introductory statement Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, andfor bereavement support groups for surviving friends and family members/loved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief (Szumilas 8: Kutcher, 2011i. contagion. Potential Outcomes II Reductions in mental health-related sequelae Increases in connectedness I Improvements in coping skills Improvements in messaging following suicide . Reductions in re-attempts I Reductions in contagion effects related to suicide I l? i I Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media} can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Therefore, responsible and safe reporting may help prevent suicide and suicide needs to be modified 3? Evidence Eurrent evidence suggests that therapeutic treatments and other approaches]for lessening harm can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, Stond?v Response Service [Stond?yi provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs (Visser, Comans, 8t Scuffham, 2014). In a study by Visser et al. [2014), StondBv clients were significantly;r less likelv to be at high risk for suicidalitvr than a suicide bereaved comparison group who had not had contact with the Steno?By program (48% and 64% respectivelv}. Additionallv, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention [versus passive approaches where survivors self~refer for services} (Cerel 8: Campbell, 2003}. Safe remitting.? messaging shout suicide. One Wav to ensure safe reporting and messaging about'suicide is to encourage news media adhere to Recommendationsfor Reporting on Suicide Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline], and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion. The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subwav, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the oualitv and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annuallv (Niederkrotenthaler 8t Sonneck, 2007). Finallv. research suggests that not only does reporting on suicide in a negative wav reporting on suicide and repetition} have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010}. Reports of individual suicidal ideation not accompanied by reports ofsuicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediatelyr following such reports [Niederkrotenthaler et al., 2010). 38 a" Comment Statement needs to be modified i Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, health care insurers and providers, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Access to Mental Health Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 39 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People At?Risk and to Lessen Harms and Prevent Future Risk. The intensity and activities of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health care, social services, and justice sectors can work collaboratively to support individuals at high- risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 40 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk grows, and monitor the effects of prevention programs and policies- -. a - CDC?suniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems and data collection sites A.E. Crosby, Ortega, et al., 2011]. Surveillance data help researchers and practitioners track changes in the burden of suicide. Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and ca use-of-death data from death certificates. is a state-based surveillance system {currently in 40 states, DC and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016). The National Electronic Injury Surveillance System-All Injury Program (NEISS-AIP) provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self-harm injuries by cause falls, poisoning, etc), age, race/ethnicity, sex, disposition {where the injured person goes when released from the Emergency Department}. In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 41 2013}. The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The National Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. Lastly, it will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. 42 In an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral wav. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantlv, this technical package extends the bounds of the typical prevention strategies to consider approaches that go bevond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. In keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. In closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 43 References Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., . . . Coventry, P. (2012). Collaborative care for depression and anxiety problems. The Cochrane Library. ArsenauIt-Lapierre, G., Kim, (2., 8L Turecki, G. (2004). diagnoses in 3225 suicides: a meta-analysis. BMC 4, K. E., Canham-Chervak, M., Black, S. A., Dada, E. 0., Millikan, A. M., S: Jones, E. H. (2012). Mental health risk factors for suicides in the US Army, 2007?8. injury Prevention, 18(6), 405-412. Ba ndura, A. (1986). Saciaifoundatians of thought and action: A social cognitive theory: Prentice-Hall, Inc. Barker, E., K?lves, K., 8i. De Leo, D. (2014). Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research, 18(3), 227-240. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi:10.1080f00048620902823714 Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukol, B. (2014). Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the Wario' Health Organization, 92(9), 641-655. Beyer, K. M., Layde, P. M., Harnberger, L. K., 8; Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49-64. doi:10.1127f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 3: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i?-ill. Bossarte, R. M., Karras, E., Lu, N., Tu, R., Stephens, B., Draper, J., 8: Kemp, J. E. (2014). Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Heaith Rep, 129(6), 516?525. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. El Ten Have, T. R. (2011]. A difference? in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol, 1 74(11), 1296-1306. doi:10.1093lajeikwr273 Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. R., Hollander, J. E., Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?5?0. Bruce, M. L, Ten Have, T. R., Reynolds C. F., Katz, 1. Schulberg, H. C., Mulsant, B. H., . . .Alexopoulos, G. S. (2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291(9), 1081?1091. Capalcli, D. M., Knoble, N. B., Shortt, J. W., Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., E: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full version?a.pdf Centers for Disease Control and Prevention. (2016). Web?Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: 44 Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. Aicohoi: Science, Poiicy and Pubiic Heaith, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 5513), 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directed Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. {2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and iife-threatening behavior, 32(3), 321-328. Cunningham, P. i. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff 28(3), w490-501. Curtin, S. C., Warner, M., St Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 45 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, 84 R. Lozano (Eds), World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 81'. Levy, S. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Doran, C. M., Ling, R., Gullestrup, J., Swannell, 5., 8: Milner, A. (2016). The Impact ofa Suicide Prevention Strategy on Reducing the Economic Cost of Suicide in the New South Wales Construction Industry. Crisis, 37(2), 121-129. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., Giles, W. H. (2001). Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA, 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1937}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding 8t G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 1897). Edwards, R. W., JumpervThurman, P., Plested, B. A., Getting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 2918,02. Escobeclo, L. (3., 81 Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 689694. Etzersdorfer, E., 81 Sonneck, G. (1993). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Felitti, V.J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks,J. S. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245-258. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., 8.: Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi:50042? 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., 8t Zhou, C. (2015). Estimated lifetime medical and workeloss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 6488), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 8: Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005?2010. Am Public Health, 105(2), 311-316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi:10.2105lAJPH2013301608 Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 46 Gilbodv, 5., Bower, P., Fletcher, J., Richards, 0., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternai medicine, 166(21), 2314-2321. Godoy Garraza, L., Walrath, C., Goldston, D. 6., Reid, H., St McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. (2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. doi:10.1111isltb.12049 Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 32(3), 338-352. Grossman, D. C., Mueller, 6. A., Riedv, C., Dowd, M. 0., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 207-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Haegerich, T. M., 8: Dahlberg, L. L. (2011). Violence as a public health risk. American Jaurnai aftifestyie Medicine, 1559822611409127. Hailey, D., Roine, R., at Ohinmaa, A. (2008). The effectiveness of telemental health applications: a review. The Canadian Joarnai of 53(11), 769278. Hambv, 5., J. (2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse Briefs in Sacioiogy. New York, NY: Springer. Han, B., Crosbv, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., 8; Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64years in the United States. Comprehensive 66, 126-186. Harris, E. C., at Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205228. Harris, E. C., 8: Barraclough, B. (1993). Excess mortality of mental disorder. Bri 173, 11?53. Harris, K. M., Carpenter, C., 3: Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499-505. Hassanian-Moghaddam, H., Sarjami, S., Kolahi, A. A., 8; Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital~treated selfnpoisoning. British ioarnai of 198(4), 309-316. Hawkins, J. 0., Catalano, R. F., 8-1 Kuklinski, M. R. (2014). Communities that ca re Encyciopedia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Haves, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 75(4), 431-456. Haves, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int;r Law 138-194. doi:10.1016ij.ijlp.2013.04.010 Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring lnternalizing Q, 26(3), 189-201. Houle, J. N., St Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am i Pubiic Heaith, 104(6), 1073-1079. 47 Hunkeler, E. M., Katon, W., Tang, L., Williams, J. Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(7536), 259-263. Inagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad201412.048 lobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8.: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12], 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. (2015). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 1-174. doi:10.15585immwr.ssE550631 Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. S., Wang, W., Toyinbo, F., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55?528. Kessler, R. (2., Demler, 0., Frank, R. (5., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515?2523. doi:10.1056lN Elk/153043266 Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 223?279. doi:10.1027l0227? 591013000136 Klomek, A. E., Sourander, A., 8; Gould, NI. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7428), 13?6. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), ?3u88. Leeb, R. T., Lewis, T., 81 Zolotor, A. J. (2011). A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of [lifestyle Medicine, 454-468. Lineberry, T. W., O'Connor, 5. S. Suicide in the US Army. Mayo Clinic Proceedings, 871-378. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 757-766. Losel, F., Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. 48 Maimon, D., Browning, C. 8., 8; Brooks-Gunn, i. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. Mark, T. L., Sherri, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8c Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. {2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention :8 Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. G., 8: Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8: Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243-253. Motto, J. A., 8: Bostrom, A. G. (2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. Author. Niederkrotenthaler, T., Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21? 41. Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 leson, M., Gerhard, T., Huang, C., S., Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Perron, S., Burrows, 5., Fournier, M., Perron, P. A., Duellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 Pirkis, 1., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 8: Cheung, T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. 49 Pitman, A., Osborn, 0., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Plested, B. A., Edwards, R. W., 31. Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rittenhouse, D. R., Fryer, G. E., Phillips, R. L., Miyoshi, T., Nielsen, C., Goodman, D. C., Grumbach, K. (2008). Impact of Title VII training programs on community health center staffing and National Health Service Corps participation. The Annais of Famiiy Medicine, 392-405. Rowhani-Rahba r, A, Simonetti, J. A, 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiai Rev, 38(1), 111-124. Rubenstein, L. V., Chaney, E. F., Ober, 5., Felker, B., Sherman, S. E., Lanto, A., at Vivell, S. (2010}. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Systems, Heaith, 28(2), 91. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emeryr Med, 17(1), 8-14. Russell, 5. T., Joyner, K. {2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Jaurnai of Pubiic Heaith, 91(8), 1276-1281. Shepard, D. S., Gurewich, D., Lwin, A. K., Reed, G. 84 Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352?362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and iife? threatening behavior, 25(1), 10?2 1. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 8: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 4959. Spoth, R. L., Guyll, M., Day, 5. X. (2002). Universal family+focused interventions in alcoholsuse disorder prevention: cost?effectiveness and cost?benefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 228. Stack, 5., 8t Wasserman, l. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and We threatening behavior, 37(1), 103-112. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. FL, 8: Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. .i Correct Heaith Care, 21(3), 222u242. Sudak, H., Maxim, K., 8! Carpenter, M. (2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. doi:10.1176fappi.ap.32.2.136 Szumilas, M., Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Pubiic Heaith, 102(1), 18-29. Toosi, M. (2015). Labor force projections to 2024: The iaborforce is growing, but siowiy. Washington, Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, G. M., 8: Crawford, M. J. {2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2016. Retrieved from 50 U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Clinician Retention: A Story of Dedication and Commitment. Retrieved from U.S. Office of the Surgeon General, at National Action Alliance for Suicide Prevention. (2012). 2012 National Strategyfor Suicide Prevention: Goals and Objectives for Action. Washington, D.C.: HHS. U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, DC. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. 5., Comans, T. A., 8: Scuffharn, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journaiof Community 42(1), 19-28. Walrath, C., Garraza, L. (3., Reid, H., Goldston, D. B., 8: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. PH.2014.302496 Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2015}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106-120. Wasserman, D., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, (3., . . . Carli, V. (2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 385(sszz), Webster?Stratton, C., Hammond, M. Treating children with early?onset conduct problems: a comparison ofchild and parent training interventions. Consuit Ciin 65(1), 93?109. Webster?Stratton, C., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283302. Webster?Stratton, C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible veers teacher and child training programs in highsrisk schools. Journal of child and 49(5), 471?488. Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., St. Anthony, J. C. (2008). The impact of two universal randomized first? and classroom interventions on voung adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, 560?23. Wilkins, N., Tsao, B., Hertz, M., Davis, H., 8L Klevens, J. (2014). Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. World Health Organization. (2013). Giobal and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and Non?Partner Sexual Violence. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurrav, M., Schmeelk-Cone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 8; Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Other Risk;r Protective Factors for Suicide Suicide Attempts or ldeation Lead Sectors1 Government (local, Unemployment benefit programs 1/ state, Federal) Strengthen economic Other income supports Businessflabor . . . . . . Housmg stabilization policies Government (local, The National Neighborhood Stabilization state, Federal) Program Coverage of mental health conditions in health insurance policies Health care Strengthen Mental Health Parity Lows 1? 1/ access to Government (state, mental health Fadera? ca re Reducing access to lethal means among persons at-risk Government (local, intervening at suicide hot spots state) 1/ Safe storage practices Public Health 1/ 1/ Create . protective Organizational pollcles and culture Busmesstabar . Together for Lg?e environments Government (local, US Air Force Suicide Prevention Program 1/ state, Fed eral) 50 Best Available Evidence Community?based policies to reduce excessive alcohol use Alcohol outlet density Government (local, state] Business/labor Promote connectedness Peer norm approaches Public Health Sources of Strength Education Communityrengagement activities Public Health Greening vacant urban spaces Government {local} Applied Suicide intervention Skills Training Social emotional learning Public Health Youth Aware of Mental Health Program Education Signs of Suicide Good Behavior Game 50"?ng Skins Parenting skill and family relationship approaches Public Health The incredible Years Education Strengthening Families 10-14 Gatekeeper training Public Health Healthcare 51 Identify and support people at-risk Screening combined with care management Best Available Evidence Henry Ford Perfect Depression Care (Pre- cursor to Zero Suicide) Healthcare Social Services Crisis Intervention Public Health National Suicide Prevention Lifeline Social Services Treatment for people at risk of suicide Healthcare improving Mood Promoting Access to Social Services Collaborative Treatment Justice Collaborative Assessment and Management Intervene to of Suicidality M3) lessen harms and revent Dialectical Behavioral Therapy 1/ future risk Attachment-Based Family Therapy Treatment to prevent re-attempts Health ca re ED Brief intervention with Follow-up Visits Social Services Active follow-up contact approaches CBT for Suicide Prevention Postvention Healthcare StandBy Response Service 52 Best Available Evidence Safe reporting following a suicide Public Health Media Guidelines Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. 53 DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development ofthis technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, Melanson, 2011). Self- directed violence may be suicide! or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is defined as a non-fotoiself-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (US. Office of the Surgeon General 8c National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 (the most recent year of available death data), suicide was responsible for 42,773 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016). In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the US (Centers for Disease Control and Prevention, 2016). Overall suicide rates increased 24% from 1999 to 2014 (Curtin, Warner, Hedegaard, 2016). Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; it is the fourth leading cause among persons in their 405 and seventh among persons in their 505. Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indian/Alaska Native and non- Hispanic White population groups. In 2014, the rates for these groups were 17.8 and 16.4 per 100,000 population, respectively (Centers for Disease Control and Prevention, 2016). Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 48% from 1999 to 2014, with steep increases seen among both males and females aged 45-64 years; (Curtin et al., 2016); Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in et al., 2012; Lineberry 8: O'Connor); workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; McIntosh et al., 2016); and lesbian, gay, bisexual, and/or queer youth, who experience increased suicidal ideation and behavior compared to their heterosexual counterparts (Kann et al., 2016; Russell Joyner, 2001). Suicides reflect only a portion of the problem (A.E. Crosby, Han, Ortega, Parks, Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le. suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (A.E. Crosby, Han, et al., 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideation) (Ferdon et al., In press). Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8i. Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one's history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or who have other risk factors noted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office ofthe Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) increases the risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts (Bossarte et al., 2014; D. P. Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, 8: Gould, 2010; Leeb, Lewis, 8: Zolotor, 2011; World Health Organization, 2013). Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other 8 instability, increases the risk for suicide and suicide attempts several fold (Bellis et al., 2014; Dube et al., 2001). The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide (Haegerich 8v: Dahlberg, 2011; Hamby 2013; Wilkins, Tsao, Hertz, Davis, Klevens, 2014). Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community (Kleiman, Riskind, Schaefer, 8; Weingarden, 2012), school (Carter, McGee, Taylor, 8: Williams, 2007), family (Maimon, Browning, Brooks-Gunn, 2010), caring adults (Capaldi, Knoble, Shortt, 8: Kim, 2012; Losel Farrington, 2012), and pro-social peers (Wyman et al., 2010) enhances resilience to suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, McIntosh, 8: Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). in an early study, Crosby and Sacks (2002) estimated that 2% ofthe U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et al (2016) found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental heaith consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (A. L. Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, McIntosh, Neimeyer, Maple, Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2003). Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegerich, Luo, 8; Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.3 billion in estimated lifetime medical and work-loss costs alone (Florence et al., 2015). Adjusting for potential under-reporting ofsuicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, 81 Silverman, 2016). The overwhelming burden of these costs were from lost productivity over the life course, with the 9 average cost per suicide being over $1.3 million {Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (US. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman E: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public le.g., business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014L Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available 10 evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cu tting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security I Housing stabilization policies Strengthen access and delivery of suicide I Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments Reduce access to lethal means among persons at? risk of suicide I Organizational policies and culture I Community-based policies to reduce excessive alcohol use Promote connectedness I Peer norm programs I Community engagement activities Teach coping and problem-solving skills I Social?emotional learning programs I Parenting skill and family relationship approaches Identify and support people at risk I Gatekeeper training I Crisis Intervention I Treatment for people at risk of suicide Treatment to prevent re-attempts Postvention I Safe reporting following a suicide Lessen harms and prevent future risk 11 It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teach Coping and Problem- Soiving strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced {Turecki, 2014). Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Oetting, Swanson, 2000; Hawkins, Catalano, 3; Kuklinski, 2014; Plested, Edwards, 8: Jumper-Thurman, 2006). These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program 12 materials, training and technical assistance] can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts}. The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide (Stack 8a Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 14 Potential Outcomes I Reductions in foreclosure rates a Reductions in eviction rates - Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal-State Unemployment lnsuronce Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss (Cylus, Glymour, 8c Avendano, 2014). An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: $7,990 per person in US. constant dollars; Cylus et al., 2014). The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5?14 weeks, 15-26 weeks, and greater than 26 weeks}, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk {Classen Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. Moreover, it wasn?t spending in general that was associated with the reduction but spending on these types of assistance. In terms of lives saved, Flavin 8L Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year {Flavin 81 Radcliff, 2009). Although this was a correlational study, 15 the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Nationai Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslv evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analvsis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014}. Another studv of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent's home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery of Suicide Ca re Rationale While most people with mental health problems do not attempt or die by suicide [Olfson, Gerhard, Huang, 8: Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {Arsenault-Lapierre, Kim, 81 Turecki, 2004; E. C. Harris Barraclough, 1997; Tyrer, Reed, El Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 8: Barraclough, 1998; World Health Organization, 2014). State-level suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shem, Bagalman, 8i Cao, 2007). Findings from the National Comorbidity Survey indicate that relatively few people in the U.S. with mental health disorders receive treatment for those conditions (Kessler et al., 2005). Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention (World Health Organization, 2014). Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care Coffey, 2007). Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: I Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity). Benefits and services covered include such things as the number of visits, co-pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. if a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others), then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. II Reduce provider shortages in underserved areas. Access to effective and state-of-the-art mental health care is largely dependent upon the training and the size of the mental health care workforce. Over 85 million Americans live in areas with an insufficient number of mental health 17 providers,- this shortage is particularly severe among low-income urban and rural communities (U.S. Department of Health and Human Services Health Resources and Services Administrations, 20163}. There are a number of ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and federal programs loan repayment programs] and expanding telemental health services. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments {see and Support People At-Hlsk, p.31}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance {National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014}. Potential Outcomes I Increases in access to mental health services 0 Increase in utilization of mental health services - Reductions in of mental illnesses and suicidality I Reductions in rates ofsuicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2005} found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013} examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated 18 with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year (Lang, 2013). Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the Notional Health Service Corps (NHSC), which offers financial incentives to attract mental/behavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps (US. Department of Health and Human Services Health Resources and Services Administrations, 2016b]. Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Telementol health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health ca re system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deaths among other outcomes (Hailey, Roine, 8: Ohinmaa, 2008}. Further, Mohr and colleagues (2008} conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face?to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy. Th us, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford healthca re system, which is a large health maintenance organization (HMO) in the state of Michigan pioneered the Perfect Depression Care program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal ofthe program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and 19 implemented coordinated continuous follow-up care system wide (C. E. Coffey, 2006). An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 8; Ahmedani, 2013). Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 (110.3 to 42.6 per 100,000; p<.04} with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 {p<.001) (M. Coffey, Coffey, Ahmedani, 2015i 20 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes [Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play, can help prevent suicide (Dahlberg 8i Krug, 2002; U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults {who comprise 42.6% of the workforce; Toosi, 2015}; among certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016), and among people in detention facilities leg. jail, prison}, to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition {Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; A.E. Crosby, Espitia-Hardeman, Ortega, 8t. Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Byck, Teplin, 2015}. Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. II Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001), and 2) that people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access {Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms-docked in a secure place in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8L Rivara, 2016; Runyan et al., 2016). 0 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments leg. residential settings). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). a Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides {Escobedo E: Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges, Wilcox, 2004). Potential Outcomes Increases in safe storage of lethal means Reductions in rates of suicide Reductions in suicide attempts 22 I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths Eviden ce The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015}. For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased fivefold lBeautrais, 2001; Beautrais et al., 2009}. Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005} found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethal Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2015] found that at post- test 76% (of the 55% of parents followed up, n=114j reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 8: Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 100,000; Mishara Martin, 2012). Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement}. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (Knox, Litts, Talcott, Feig, Caine, 2003). Using a time-series design to examine the impact of the program on various violence?related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide (Knox et al., 2003}.The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death {Knox et al., 2003). A longitudinal assessment ofthe program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post?launch) found significantly lower rates of suicide after the program was launched than before (Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation in that year than in the other years (Knox et al., 2010). Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests that organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behavior/suicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates 24 dropped from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year (Hayes, 1995). Other similar programs have seen declines in suicide both in the United States and internationally (Barker, K?lves, 8: De Leo, 2014]. Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density, specifically, suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that the density of bars, specifically, is inversely related to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, Remer, 2009). 25 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951}. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 3; Krug, 2002], for instance between individuals peers, neighbors, co-workers), families, schools, neighborhoods, workplace, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, 8; Zhu, 2013]. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest the pattern is towards a positive association between social capital measured by social trust, community/neighborhood engagement, and improved mental health. Connectedness and social capital together can serve to protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, personal value and worth all of which helps individuals to build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009). Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 26 enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors I Increases in referrals for youth in distressed I Increases help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (E metropolitan, 12 rural), Wyman et al. (2010) found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010}. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. Other benefits included reductions in firearm assaults and vandalism {Branas et al., 2011). 27 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004}. Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on develooing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, &Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors Ii.e., depression, anxiety, conduct problems, substance abuse) 0 Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8: Hunter, 2010). Social emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wasserman et al., 2014). In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts amongthe YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group absolute risk fell by 0.50% and relative risk fell by 49.6% (Wasserman et al., 2014}. Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008]. Two cohorts of youths participated in the program in 1985-86 and 1985?87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation 29 and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect of the program was consistent for suicidal ideation regardless of whether baseline covariates were included. The 636 effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of (386 students, neither suicidal ideation nor suicide attempts were significantly different between 636 and the control interventions (Wilcox et al., 2008}. The researchers believed this may have been due to a lack of implementation fidelity. 636 was also found to be associated with reduced risk of later substance abuse, a risk factor for suicide (Kellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide riskfactors in youth by improving protective factors such as responsive and positive parent-teacher-child interactions and relationships, emotion self?regulation and social competence {all protective factors for suicide) (Herman et al., 2011). The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0? program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster-Stratton, Jamila Reid, 8: Stoolmiller, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; C. Webster-Stratton 8: Hammond, 1997; C. Webster-Stratton, Reid, 8; Hammond, 2001). The program demonstrated greater benefits as the dosage of the intervention increased (Herman et al., 2011). Additionally, Strengthening Families 10?14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide (Spoth, Guyll, 8: Day, 2002). Strengthening has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 30 Identify and Support People At-Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; Curtin et al., 2016; Kann et al., 2016; Lineberry 8i. O'Connor; Russell 8L Joyner, 2001}. Sapporting these groups requires proactive case finding and effective response, crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible internet-based services when appropriate} and engaging people in evidence?based care leg. through such measures as collaborative treatment), remain key challenges. Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services. Approaches The following approaches focus on identifying and supporting people at increased risk. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis (or a friend or family member of someone at-risk) to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Treatment for people at-risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors 31 with problem-solving and emotion regulation. Treatment usually takes place in a one on one or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/'or integrated care linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk {Archer et al., 2012; Bruce et al., 2004; Gilbody, Bower, Fletcher, Richards, 8: Sutton, 2005}. Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296?2594: reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (Inagaki et al., 2015) Potential Outcomes Reductions in suicide attempts Reductions in suicide deaths Reductions in of mental illnesses and suicidal ideation Reductions in mental health-related sequelae Reductions in re-attempts Increases in connectedness Improvements in coping skills Increases in identification of individuals at-risk for suicidal behavior Increases in treatment engagement by at-risk individuals Increases in community members trained to identify at-risk individuals Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to 32 identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman (2013] evaluated the training across the Notionni Suicide Prevention Lifeline network of hotlines over the period 2008- 2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, the researchers found that compared to callers who spoke to counselors that received the usual care training, individuals who spoke with counselors without training in were significantly more likely to feel depressed, suicidal, more overwhelmed, and less hopeful by the end of their call to the hotline compared to those with training in Counselors trained in were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASLST did not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. Gatekeeper training has also been a primary component of the Gorret Lee Smith Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10-24 in counties implementing GLS trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation {Walrath, Garraza, Reid, Goldston, 8L McKeon, 2015). This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 23? deaths in the age group, between 2007 and 2010. Counties implementing GLS program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GLS program than did similar counties that did not implement GLS activities {4.9 fewer attempts per 1000 youths,- Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015). More than 79,000 suicide attempts may have been prevented during the period examined, following implementation of the GL5 program. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeline to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow?up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1) to follow-up (time 3). Between time 2 (end of the call) to time 3, the effect remained for pain and 33 hopelessness, but was not significant for intent to die (Gould, Kalafat, Harrismunfakh, 8L Kleinman, 200?}. Treatment for people at?risk of suicide. The improving Mood Promoting Access to Coiiaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase] by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up (Hunkeler et al., 2006; Unutzer et al., 2006] relative to patients who received care as usual. Coiiaborative Assessment and Management of Saicidaiity (CAMS), is a therapeutic approach for suicide-specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS been tested and supported in 6 correlational studies (Jobes, 2012}, in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual (intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow-up (Comtois et al., 2011). Other examples include Diaiecticai Behavioral Therapy (DST) and Attachment?Based Family Therapy (ABFT). a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBT were half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self?injurious acts combined [Linehan et al., 2006}. 34 ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%) and at 24 weeks (82.1% vs. 46.2%) (Diamond et al., 2010). The Veterans Affairs Transiating initiatives for Depression into E??ective Soiutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 20% of primary care patients (Rubenstein et al., 2010). also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively (Rubenstein et al., 2010). Treatment to prevent re?attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brie}r intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific time line for up to 18?months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with 9 follow-up visits over 18-months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively) (Fleischmann et al., 2008). Another example of treatment to prevent re-attempts involves active foiiow-ap contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years). In a meta-analysis conducted by Inagaki et al. (2015), interventions to prevent repeat suicide attempts in patients 35 admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of the post-crisis suicide prevention long-term follow?up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts {Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). 36 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world. Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide (Pitman, Osborn, King, EL Erla ngsen, 2014}. Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion {Etzersdorfer 3t Sonneck, 1998; Niederkrotenthaler E: Sonneck, 2007}. While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care forthe bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions} and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief (Szumilas Kutcher, 2011). 0 Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are both inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline}, and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion. Potential Outcomes I Reductions in ideation/attempts 37 I Reductions in distress I Increases in treatment seeking - Improvements in reporting following suicide - Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others? suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service {5tondBy}, provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs {Vissen Comans, 8L Scuffham, 2014). In a study by Visser et al. (2014), StondBy clients were significantly less likely to be at high risk for suicidality {suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) {Cerel 8: Campbell, 2008). Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendations for Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually (Niederkrotenthaler 8: Sonneck, 2007). Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010). Reports of individual suicidal ideation not accompanied by reports ofsuicide or suicide attempts, along with reports describing a "mastery? of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports {Niederkrotenthaler et al., 2010). 38 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 39 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 40 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andfor medicolegal standards (A.E. Crosby, Ortega, et al., 2011). uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems A.E. Crosby, Ortega, et al., 2011). Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. Notional Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of?death data from death certificates. is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016]. Data from state and local Child Death Review teams and Suicide Death Review Teams {which are in a few states] offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic lnj'ury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department}. In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade 41 students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The l?llotiomtzlI Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non?medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groUps at increased risk. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies (before risk occurs) and community?level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, 42 mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 43 References Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, D., Gask, L., . . . Coventry, P. (2012). Collaborative care for depression and anxiety problems. The Cochrane Library. ArsenauIt-Lapierre, G., Kim, (2., 8L Turecki, G. (2004). diagnoses in 3225 suicides: a meta-analysis. BMC 4, K. E., Canham-Chervak, M., Black, S. A., Dada, E. 0., Millikan, A. M., S: Jones, E. H. (2012). Mental health risk factors for suicides in the US Army, 2007?8. injury Prevention, 18(6), 405-412. Ba ndura, A. (1986). Saciaifoundatians of thought and action: A social cognitive theory: Prentice-Hall, Inc. Barker, E., K?lves, K., 8i. De Leo, D. (2014). Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research, 18(3), 227-240. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. doi:10.1080f00048620902823714 Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukol, B. (2014). Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the Wario' Health Organization, 92(9), 641-655. Beyer, K. M., Layde, P. M., Harnberger, L. K., 8; Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49-64. doi:10.1127f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 3: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppil, i?-ill. Bossarte, R. M., Karras, E., Lu, N., Tu, R., Stephens, B., Draper, J., 8: Kemp, J. E. (2014). Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Heaith Rep, 129(6), 516?525. Branas, C. C., Cheney, R. A., MacDonald, J. M., Tam, V. W., Jackson, T. El Ten Have, T. R. (2011]. A difference? in-differences analysis of health, safety, and greening vacant urban space. Am Epidemiol, 1 74(11), 1296-1306. doi:10.1093lajeikwr273 Brown, G. K., Ten Have, T., Henriques, G. R., Xie, S. R., Hollander, J. E., Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?5?0. Bruce, M. L, Ten Have, T. R., Reynolds C. F., Katz, 1. Schulberg, H. C., Mulsant, B. H., . . .Alexopoulos, G. S. (2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291(9), 1081?1091. Capalcli, D. M., Knoble, N. B., Shortt, J. W., Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., E: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full version?a.pdf Centers for Disease Control and Prevention. (2016). Web?Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: 44 Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. Aicohoi: Science, Poiicy and Pubiic Heaith, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 5513), 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directed Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. {2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and iife-threatening behavior, 32(3), 321-328. Cunningham, P. i. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff 28(3), w490-501. Curtin, S. C., Warner, M., St Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 45 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, at. R. Lozano (Eds), World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., vlilintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 3-: Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., Giles, W. H. {2001). Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA, 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1937}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 1897). Edwards, R. W., Jumper-Thurma n, P., Plested, B. A., Getting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 291-302. Escobedo, L. (5., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accla' Anal Prev, 34(5), 689-694. Etzersdorfer, E., Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Felitti, J., Anda, R. F., Nordenberg, 0., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many ofthe leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245?258. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., 8: Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., St Zhou, C. (2015). Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 64(38), 1074?1072. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 81 Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 3114316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi:10.2105lAJPH.2013.301608 Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Gilbody, 5., Bower, P., Fletcher, J., Richards, D., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternal medicine, 165(21), 2314-2321. 46 Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 3?07-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heoith Reports, 95(5), 411-421. Haegerich, T. M., 8: Dahlberg, L. L. (2011). Violence as a public health risk. American Journai oftifestyie Medicine, 1559827611409127. Hailey, D., Roine, H., 81 Ohinmaa, A. (2008}. The effectiveness of telemental health applications: a review. The Canadian Journai of 53(11), 769-278. Hamby, 5., J. [2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse Briefs in Socioiogy. New York, NY: Springer. Han, E., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 84 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?E4years in the United States. Comprehensive 66, 176486. Harris, E. (2., 8L Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205-228. Harris, E. (2., 8L Barraclough, B. (1998). Excess mortality of mental disorder. Br.i 173, 11?53. Harris, K. M, Carpenter, (2., 8t Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian?Moghaddam, H., Sarjami, S., Kolahi, A. A., Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital~treated selfnpoisoning. British Journai of 198(4), 309-316. Hawkins, J. D., Catalano, R. F., 8t. Kuklinski, M. R. (2014). Communities that ca re Encyciopeo?ia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Hayes, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 25(4), 431-456. Hayes, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int} Law 188?194. Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's [Io-Occurring Internalizing 0, 26(3), 189-201. Houle, J. N., 8t Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am .i Puhiic Heaith, 104(6), 1073-1029. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-263. 47 lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 6140-653. doi:10.1111ij.1943- 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12), 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. {2016). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 31?174. Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, FL, Giesbrecht, N., 3: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. (3., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. (2., Demler, 0., Frank, R. 6., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. EnglJ Med, 352(24), 2515?2523. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi:10.102710227~ 5910/a000136 Klomek, A. E., Sourander, A., at Gould, M. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross?sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 3: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7423), 1376. Knox, K. L., Pflanz, S., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2452-2453. doi:10.2105lajph.2009.159871 Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 73-88. doi:10.1002ihec.1816 Leeb, R. T., Lewis, T., 82 Zolotor, A. J. (2011}. A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal ofth?estyle Medicine, 4454-468. Lineberry, T. W., 3; O'Connor, 5.5. Suicide in the US Army. Mayo Clinic Proceedings, 87(9), 821-373. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 75?-766. Losel, F., St Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Maimon, 0., Browning, C. R., 8: Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. 48 Mark, T. L., Shern, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., 8; Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention 3: Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8L Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8: Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243-253. Motto, J. A., 3: Bostrorn, A. G. (2001}. A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. (2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprint for Workplace Suicide Prevention. Washington, DC. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. Author. Niederkrotenthaler, T., 8r Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust Z.l 41. doi:10.1080l00043620701266680 Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 Olfson, M., Gerhard, T., Huang, C., 5., EL Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193?199. Perron, S., Burrows, S., Fournier, M., Perron, P. A., 8-1 Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105iajph.2012.301089 Pirkis, J., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 84 Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. 49 Plested, B. A., Edwards, R. W., Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rowhani-Rahba r, A., Simonetti, J. A., Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. Rubenstein, L. Chaney, E. F., Ober, 5., Felker, B., Sherman, 5. E., Lanto, A., 8; Viva?, 5. (2010]. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Systems, Heaith, 28(2), 91. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West} Emerg Med, 17(1), 8-14. Russell, 5. T., 8L Joyner, K. (2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Journai of Pubiic Heaith, 91(8), 1276-1281. Shepard, D. 5., Gurewich, D., Lwin, A. K., Reed, 6. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and We? threatening behavior, 25(1), 10?21. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 3: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M, 81 Day, 5. K. (2002). Universal family~focused interventions in alcohol~use disorder prevention: cost-effectiveness and cost?benefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 223. Stack, 5., 8t Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and iife? threatening behavior, 37(1), 103-112. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. R., Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Correct Heaith Care, 21(3), 222?242. Sudak, H., Maxim, K., Carpenter, M. {2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. doi210.1176/appi.ap.32.2.136 Szumilas, M., 8; Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can .i Pubiic Heaith, 102(1), 13?29. Toosi, M. (2015). Laborforce projections to 2024: The iaborforce is growing, but siowiy. Washington, D.C.: Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, G. M., 8: Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-725. U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2016. Retrieved from U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Ciinician Retention: A Story of Dedication and Commitment. Retrieved from U.S. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 Nationai Strategyfor Suicide Prevention: Goais and Objectives for Action. Washington, DC: HHS. 50 U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primarv care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Ga rraza, L. (3., Reid, H., Goldston, D. B., 8-: McKeon, R. (2015). impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 105-120. Wasserman, 0., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, G., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 1536-1544. Webster-Stratton, C., Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 55(1), 93-109. Webster-Stratton, C., Reid, M. J., 8: Hammond, M. (2001]. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283?302. Webster?Stratton, C. H., Reid, M. J., 81 Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191?203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible veers teacher and child training programs in high?risk schools. Journal of child and 49(5), 471?488. Wilcox, H. (2., Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., St Anthony, J. C. (2008). The impact of two universal randomized first? and second~grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, S6043. Wilkins, N., Tsao, B., Hertz, M., Davis, R., St Klevens, J. (2014). Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. World Health Organization. (2013). Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and NonaPartner Sexual Violence. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurrav, M., Schmeelkaone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 81 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment benefit programs v? 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization v, state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity Laws 1/ 1/ Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) ?We? 0? National Health Service Corps (NHSC) suicide care Healthcare Telemental health (TMH) Social services Safer sunclde care through systems change Henry Ford Perfect Depression Care {Pre- v, V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt intervening at suicide hot spots Government (local, state) Create - Safe storage practices . protective Public Health environments Emergency Department Counseling on v, Organizational policies and culture 50 Best Available Evidence Together for Life US Air Force Suicide Prevention Program Correctional suicide prevention Business/Labor Justice Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Alcohol outlet density 1/ Government (local, state) Business/labor Promote connectedness Peer norm programs Public Health Sources ofStrength Education Communit en a ement activities 5 Public Health Greening vacant urban spaces Government {local Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mental Health Program Education Good Behavior Game Parenting skill and family relationship approaches Public Health The incredible Years Education 51 Strengthening Families 10-14 Best Available Evidence Identify and at-risk support people Gatekeeper training Public Health Applied Suicide intervention Skilis Training Healthcare Garret Lee Smith Federal Grant Program v? Crisis Intervention Public Health Soc'al Services National Suicide Prevention Lifeline v? I Treatment for people at risk of suicide improving Mood - Promoting Access to 1/ Collaborative Treatment Collaborative Assessment and Management Healthca re of Suicidality (CAMS) Dialectical Behavioral Therapy (DST) Somal Servlces Justice Attachment-Based Family Therapy v? Translating initiatives for Depression into Ejj?ective Solutions project Treatment to prevent re-attempts ED Brief intervention with Follow-up Visits Health ca re Active follow-up contact approaches Social services CBTfar Suicide Prevention 52 Best Available Evidence Postvention Lessen harms StondBv Response Service 1/ 1/ ea care and prevent future risk Safe reporting following a suicide Public Health 1/ Media Gurdelmes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development ofthis technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government could support the implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called self-directed violence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury (Crosby, Ortega, 8: Melanson, 2011). Self-directed violence may be suicidal or non-suicidal in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. I Suicide attempt is defined as a non-fatal self-directed and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). in 2015 (the most recent year of available death data), suicide was responsible for 492.123 deaths in the U.S., which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention, 2016). In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1975 in the U.S (Centers for Disease Control and Prevention, 2016}. Overall suicide rates increased 24% from 1999 to 2014 (Curtin, Warner, 81 Hedegaard, 2016), with the inclusion of 2015 data, that percentage has risen to more than 25%. Suicide is a problem throughout the life span; it is the second leading cause of death among those aged 10?34 years; the fourth leading cause among persons in their 405, and seventh leading cause among persons in their 505. Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among non-Hispanic American Indiaanlaska Native and non- Hispanic White population groups. In 2014, the rates for these groups were 17.8 and 16.4 per 100,000 population, respectively (Centers for Disease Control and Prevention, 2016). Other population groups disproportionately impacted by suicide include middle-aged adults (whose rates increased 48% from 1999 to 2014, with steep increases seen among both males and females aged 45-64 years; (Curtin et al., 2016); Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the first time in decades; et al., 2012; Lineberry 8: O'Connor, 2012); workers in certain occupational groups protective service occupations; workers in farming, fishing, and forestry; McIntosh et al., 2016}; and lesbian, gay, bisexual, andfor queer (LGBQ) youth, who experience increased suicidal ideation and behavior compared to their heterosexual counterparts (Karin et al., 2016; Russell 8: Joyner, 2001). Suicides reflect only a portion of the problem (Crosby, Han, Ortega, Parks, 8E. Gfroerer, 2011). Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all (Crosby, Han, Ortega, Parks, 3i. Gfroerer, 2011). For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide (Le. ideation} (David-Ferdon, et al., 2016) Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal influences that interact with one another, often over time. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg Krug, 2002). Risk and protective factors for suicide exist at each level. For example, risk factors include: 0 individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants 0 Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, familyj'loved one?s history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications} Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other noted risk factors, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office ofthe Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk ofdepression, post-traumatic stress disorder anxiety, suicide, and suicide attempts (Bossarte et al., 2014; Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, Gould, 2010; Leeb, Lewis, 81. Zolotor, 2011; World Health Organization, 2013). Women exposed to 8 partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is associated with increased risk for suicide and suicide attempts several fold (Bellis et al., 2014; Dube et al., 2001). The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress ?factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide (Haegerich Dahlberg, 2011; Hamby 2013; Wilkins, Tsao, Hertz, Davis, 8: Klevens, 2014). Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community (Kleiman, Riskind, Schaefer, 8: Weingarden, 2012), school (Carter, McGee, Taylor, 8.1 Williams, 2007), family (Maimon, Browning, Brooks-Gunn, 2010), caring adults (Capaldi, Knoble, Shortt, Kim, 2012; Losel 8: Farrington, 2012), and pro-social peers (Wyman et al., 2010) can enhance resilience and help reduce risk for suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, 8: Dunne- Maxim, 1987; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015). in an early study, Crosby and Sacks (2002) estimated that 7% of the US. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. in a more recent study, in one state, Cerel et al. (2016) found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide and/or having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss) is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt (Chapman 8: Dixon-Gordon, 2007). Similarly, survivors of a loved one's suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, Mortimer-Stephens, 2004), stigma, depression, anxiety, post-traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, Neimeyer, Maple, 8: Marshall, 2014; Sudak, Maxim, 8: Carpenter, 2008). Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegerich, Luo, 8: Zhou, 2015). The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work-loss costs alone (Florence et al., 2015). Adjusting for potential under-reporting ofsuicide and drawing upon health expenditures per capita, GDP 9 per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, 8L Silverman, 2016). The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million (Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (U.S. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman 8: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012). Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014). Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design] evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context ifthe program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on 10 suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cutting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security - Housing stabilization policies Strengthen access and delivery of suicide I Coverage of mental health conditions in health care insurance policies I Reduce provider shortages in underserved areas Ir Safer suicide care through systems change Create protective environments Reduce access to lethal means among persons at? risk of suicide - Organizational policies and culture II Community-based policies to reduce excessive alcohol use Promote connectedness Peer norm programs I Community engagement activities Teach coping and problem-solving skills - Social~emotional learning programs I Parenting skill and family relationship approaches identify and support people at risk I Gatekeeper training I Crisis Intervention I Treatment for people at risk of suicide 11 0 Treatment to prevent resattempts Lessen harms and prevent future risk I- Postvention - Safe reporting following a suicide It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teach Coping and Problem? Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, racefethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals' coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced {Turecki, 2014). Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Oetting, Swanson, 2000; Hawkins, Catalano, 8: Kuklinski, 2014; Plested, Edwards, Jumper-Thurman, 2006). These planning processes engage and guide community 12 stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation le.g., availability of program materials, training and technical assistance} can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks), but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts). The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide (Stack 8a Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 14 Potential Outcomes I Reductions in foreclosure rates a Reductions in eviction rates - Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal-State Unemployment lnsuronce Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss (Cylus, Glymour, 8c Avendano, 2014). An examination of variations in unemployment benefit programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: $7,990 per person in US. constant dollars; Cylus et al., 2014). The effects of unemployment benefit programs were also consistent by sex and age group. Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5?14 weeks, 15-26 weeks, and greater than 26 weeks}, and job losses found that the duration of unemployment, as opposed to just the loss ofjob, predicted suicide risk {Classen 81 Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their financial allocations, but also in their duration. Other measures to strengthen household financial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance} have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families TANF) and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors le.g., residential mobility, divorce rate, unemployment rate} at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin 81 Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year (Flavin 8f. Radcliff, 2009). Although this was a correlational study, the results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and 15 increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Nationai Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle 8L Light, 2014). Another study of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began) to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent?s home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery ofSuicide Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, El. Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {Arsenault-Lapierre, Kim, 34 Turecki, 2004; Harris 8; Barraclough, 1997; Tyrer, Reed, Crawford, 2015), previous research indicates that mental illness is an important risk factor for suicide {Harris Barraclough, 1998; World Health Organization, 2014}. Stateslevel suicide rates have also been found to be correlated with general mental health measures such as depression (Lang, 2013; Mark, Shem, Bagalman, Si. Cao, 200?}. Findings from the National Comorbidity Survey indicate that relatively few people in the US. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention {World Health Organization, 2014). Additionally. research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care (Coffey, 200?}. Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: Coverage of mental health conditions in health insurance policies. Federal state laws} I include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns li.e., mental health parity}. Benefits and services covered include such things as the number of visits, co?pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the Comment IAI: Double check and make sure that this will not need an ACA review state parity law. If a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others}, then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. I Reduce provider shortages in underserved areas. Access to effective and state?of?the?art mental health care is largely dependent upon the training and the size ofthe mental health care th workforce. Over 35 million Americans live in areas with an insufficient number of mental heal 1? providers,- this shortage is particularly severe among low-income urban and rural communities (U.S. Department of Health and Human Services Health Resources and Services Administrations, 2016a). There are a range of ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and federal programs loan repayment programs] and expanding telemental health services [the use of telephone, video and web~based technology to provide or care at a distance]. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be delivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments {see identify and Support People At-RlSk, p.31), continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance {National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014}. Potential Outcomes I Increases in access to mental health services I Increase in utilization of mental health services I Reductions in of mental illnesses and suicidality 0 Reductions in rates ofsuicide attempts 0 Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and service utilization. Using data from this survey, Harris, Carpenter, and Bao (2006) found that 12 months after states enacted mental health parity lows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013} examined state mental health laws and suicide rates 18 between 1990 and 2004 and found that mental health parity laws, specifically, were associated with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year (Lang, 2013). Reduce provider shortages in underserved areas. One example ofa program to improve access to mental health care providers is the Notionoi Heoith Service Corps (NHSC), which offers financial incentives to attract mentalfbehavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps (US. Department of Health and Human Services Health Resources and Services Administrations, 2016b). Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Telementol health services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities) to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deaths among other outcomes (Hailey, Roine, Ohinmaa, 2008). Further, Mohr and colleagues (2008) conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face-to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy. Thus, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Healthcare System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered the Perfect Depression Core program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Core was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, 19 timeliness, efficiency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow?up care system wide (Coffey, 2006). An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% (Coffey, 2006; Coffey, Coffey, Ahmedani, 2013). Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 (110.3 to 47.6 per 100,000 population; p<.04} with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per 100,000 (p<.001) (Coffey, Coffey, Ahmedani, 2015). 20 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes (Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide (Dahlberg 8i. Krug, 2002; U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults {who comprise 42.6% of the workforce; Toosi, 2015}; among certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016), and among people in detention facilities leg. jail, prison}, to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition {Beautrais, Gibb, Fergusson, Horwood, 8i Larkin, 2009; Crosby, Espitia-Hardeman, Ortega, Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i. Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Byck, Teplin, 2015). Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. II Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001), and 2) people tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007; Tip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms-docked in a secure place in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8L Rivara, 2016; Runyan et al., 2016). 0 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments leg. residential settings). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). a Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides {Escobedo E: Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges, Wilcox, 2004). Potential Outcomes Increases in safe storage of lethal means Reductions in rates of suicide Reductions in suicide attempts 22 I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths Eviden ce The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspat interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased five- fold (Beautrais, 2001; Beautrais et al., 2009). Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethal Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2015) found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking (Mishara 8: Martin, 2012). Police suicides were tracked over 12 years and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 73.9% to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city (29.0 per 100,000; Mishara 8; Martin, 2012}. Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates every suicide outcomes measurement). The program represents a fundamental shift from viewing suicide and mental illness solely as medical problem and instead sees them as larger service- wide problems impacting the whole community (Knox, Litts, Talcott, Feig, El Caine, 2003]. Using a time-series design to examine the impact of the program on various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide {Knox et al., 2003}.The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%, respectively), homicide and accidental death (Knox et al., 2003). A longitudinal assessment ofthe program over the period 1981 to 2008 (15 years before the 1997 launch of the program and 11 years post-launch} found significantly lower rates of suicide after the program was launched than before launch (Knox et al., 2010). These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of all program components across all Air Force installations in that year than in the other years (Knox et al., 2010i Finally, while the evidence is still being built for suicide prevention in carrectionai facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varying levels of observation, safe physical environment, emergency response protocols, notification of suicidal behaviorfsuicide through the chain of command, and critical incident stress debriefing and death review can potentially reduce suicide. When these policies and 24 practices were implemented across 11 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year (Hayes, 1995). Other similar programs have seen declines in suicide both in the United States and internationally (Barker, Kolves, 8; De Leo, 2014). Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use of alcohol exist, several studies on alcohol outlet density and suicide risk factors, such as violence, {Gruenewald Remer, 2006; Lipton 8: Gruenewald, 2002; Rush, Gliksman, Brook, 1986) suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes indicated that greater density of bars, specifically, is related to greater suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, 8L Remer, 2009). 25 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 1897 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, 1897/1951}. Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009). Social connections can be formed within and between multiple levels ofthe social ecology {Dahlberg 8r. Krug, 2002], for instance between individuals peers, neighbors, co?workers), families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, 8: Laud, 2015; Muennig, Cohen, Palmer, 8; Zhu, 2013]. Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital measured by social trust, community/neighborhood engagement, and improved mental health. Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009]. Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. I Community engagement activities. Community engagement is an aspect of social capital. Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 26 enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. Potential Outcomes I Reductions in maladaptive coping attitudes and behaviors I Increases in healthy coping attitudes and behaviors I Increases in referrals for youth in distressed I Increases help-seeking behaviors I Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive benefits of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools (E metropolitan, 12 rural), Wyman et al. (2010) found that the program improved adaptive norms regarding suicide among peer leaders, connectedness to adults, and school engagement. Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths, particularly among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders {Wyman et al., 2010}. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2003. Local residents and community members worked together to green 4,436 lots (or 7.8 million square feet) in 4 areas ofthe city. Researchers found significant reductions in community residents? self-reported level of stress, which is a risk factor for suicide, and engagement in more physical exercise, a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other benefits, including reductions in firearm assaults and vandalism (Branas et al., 2011}. 27 Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004}. Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on develooing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, &Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes I Reductions in suicide attempts and suicide ideation Reductions in suicide risk behaviors Ii.e., depression, anxiety, conduct problems, substance abuse) 0 Improvements in help-seeking behavior I Improvements in social competence and emotional regulation skills I Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. S. Knox, Burkhart, 8: Hunter, 2010). Social emotional learning programs. The Youth Aware of Mental Health Program (mm) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wasserman et al., 2014). In a cluster-randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were significantly less likely to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts amongthe YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the YAM group compared to 11 in the control group. Additionally, related to severe suicide ideation, in the YAM group relative risk fell by 49.6% [Wasserman et al., 2014]. Another example is the Good Behavior Game (636), which is a classroom-based program for elementary school children aged 6-10. The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior (Wilcox et al., 2008]. Two cohorts of youths participated in the program in 1985-86 and 1985?87 school years when they were in the first and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide-related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 636 when they were in the first grade reported half the adjusted odds of suicidal ideation 29 ,{Cnmment How so? If Comment IAI: in the last version you had added ?and pointed to the need for of the program was consistent for suicidal ideation regardless of whether baseline covariates GBG to be delivered with precision, consistency, and teacher support." It were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of 636 students, neither suicidal ideation nor i" could help to put something like this back 5' in at the end of this sentence. suicide attempts were significantly different between 636 and the control interventions 5} he researchers believed this may have been due to a lack of :f and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect Comment One or both cohort? i? Please confirm. Comment IAI: Specific outcomes or all outcomes? {Wilcox et al., 2008). [i "1 Jun. JL. Herman et al. indicate the ideal number i i Comment IAI: Number of sessions? no of from 9 to 20}? also found to be associated with E'educed risk of later substance abus and other suicide risk factors among the first cohort of students. Results for the second cohort were generally smaller but in the desired {Kellam et al., 2008). I Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with ii suicide. For example, the incredibie Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, if two important suicide risk factors in youth by improving protective factors such as responsive ii and positive parent-teacher-child interactions and relationships, emotion self-regulation and :i social competence [all protective factors for suicide} (Herman et al., 2011]. The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals}. Several studies have demonstrated ii the effect of the it? program on reducing internalizing such as anxiety and ii depression, and child condUct problems (Webster?Stratton, Reid, 8: Beauchaine, 2011; Webster-Stratton, iamila Reid, 8: Stoolmiiler, 2003). The program is also associated with i improved problem?solving and conflict management; these skills were maintained at 1-year ,il' follow-up (Reid, Webster-Stratton, 8: Hammond, 2003; Webster-Stratton 8i. Hammond, 199?; Wehster?Stratton, Reid, Er Hammond, 2001). program demonstratedifr greater benefitslin mother-rated child internalizing compared to the waitlistedg so when pers- os shims and teacher somerset: wsrs. included-mitoses? et al., 2011). Additionally, Strengthening 10-14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths? interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent?child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance 30 abuse, two important risk factors for suicide (Spoth, Guy?, 8: Day, 2002). Strengthening Families has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families (Spoth et al., 2002). 31 ldentil?y and Support E?eople Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have previously attempted suicide; iv'eterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status and members of certain racial and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; Curtin et al., 2015; Kann et al., 2015; Lineberry 8t O'Connor, 2-312; Husseil 8L Joyner, 2001}. Supporting these at-risk groups requires proactive case finding and effective response, crisis intervention, and evidence?ha sed treatment. Finding optimal ways of identifying at-risk individuals, costomizing services to make them more accessible Internet?based services when appropriate} and engaging people in evidence?based care through such measures as collaborative treatment], remain key chailenges. Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face sociai and economic issues that can adversely affect their ability to access supportive services. Approaches The following approaches focus on identifying and supporting people at increased risk. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typically by connecting a person in crisis [or a friend or family member of someone at~risi<} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feeiings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. I Treatment for people at-risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors 32 1 I ,x'l Comment Suggest: At?Rlsk People I Comment IAI: Prefer to keep as i5? with problem-solving and emotion regulation. Treatment usually takes place in a one-on-one or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/'or integrated care ie.g., linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk {Archer et al., 2012; Bruce et al., 2004; Gilbody, Bower, Fletcher, Richards, 8: Sutton, 2005}. Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296?2594: reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (Inagaki et al., 2015) Potential Outcomes Reductions in suicide attempts Reductions in suicide deaths Reductions in of mental illnesses and suicidal ideation Reductions in mental health-related sequelae Reductions in re-attempts Increases in connectedness Improvements in coping skills Increases in identification of individuals at-risk for suicidal behavior Increases in treatment engagement by at-risk individuals Increases in community members trained to identify at-risk individuals Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Skilis Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to 33 identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, Kleinman (2013) evaluated the training across the Notionoi Suicide Prevention Lifeiine network of hotlines over the period 2008- 2009. Using data from 1,410 suicidal individuals who called 1? Lifeline centers, the researchers found that callers who spoke with ASST-trained counselors were significantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end of their call, compared to callers who spoke to counselors. Counselors trained in also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in did not result in more comprehensive suicide risk assessments than usual care training {Gould et al., 2013}. Gatekeeper training has also been a primary component of the Garret Lee Smith (GL5) Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10-24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantly lower youth suicide rates one year following the training implementation [Walrath, Garraza, Held, Goldston, EL McKeon, 20151. This finding equates to a decrease of 1 suicide death per 100,000 among youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had significantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities {4.9 fewer attempts per 1000 youths; Godoy Garraza, Walrath, Goldston, Reid, McKeon, 2015]. More than 79,000 suicide attempts may have been prevented during the period following implementation of the GL5 program. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention Lifeiine to prevent suicide, 1,035 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-Up assessment between 1 and 52 days [meanzlELS days) after the initial assessment. Researchers found that over half of the initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow?up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call (time 1] to follow-up [time 3). etween time 2 {end of the call) to time 3, the effect remained for pain and} hopelessness, but was not signi?cant for intent to diei, suggesting that greater effort at outreach 34 Comment IAI: This text is somewhat confusing. The fact that the effect for intent to die did not "remain? signi?cant suggest that it went back up to what it was when they initiated the call. Also this ends on a disconcerting note. in the other places where you had an important null result you included an idea about why and what was needed ie.g., Can you . Ldo that here too? during and following the call is needed for the callers with high levels of suicide intent EGouId, Kalafat, Harrismunfakh, a Kleinman, 2007} ?ip-[Comment I AI: ?3?.th without Treatment for people at-risk of suicide. The improving Mood Promoting Access to Collaborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekly during an acute phase and during continuation phase) by a depression care manager {Hunkeier et al., 2006). The program has been shown to significantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24?months of follow-up (Hunkeler et al., 2006; Unutzer et al., 2006] relative to patients who received care as usual. Coiioborotive Assessment and Management ofSuicidoiity (CAMS), is a therapeutic approach for suicide?specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient~specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS was tested and supported in correlational studies (lobes, 2012]. in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual {intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found bettertreatment retention among the CAMS group and significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month foilow~up [Comtois et at, 2011). Other examples include Dialectical Behavioral Therapy and Attachment?Based Fomiiy Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues. The components of BET include individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at two-year follow-up than women receiving community treatment [23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self?injurious acts combined (Linehan et al., 2006). 35 ?(Comment Fix formatting changing the margins to non-justified. i ABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%) and at 24 weeks (82.1% vs. 46.2%) (Diamond et al., 2010). The Veterans Affairs Transiating initiatives for Depression into E??ective Soiutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 20% of primary care patients (Rubenstein et al., 2010]. also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively (Rubenstein et al., 2010). Treatment to prevent re?attempts. Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brie}r intervention with Foiiow-up Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 18 months). Follow-up contacts are either conducted by phone or through home visits according to a specific timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively) (Fleischmann et al., 2008}. Another example of treatment to prevent re-attempts involves active foiiow-ap contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years). In a meta-analysis conducted by Inagaki et al. (2015), interventions to prevent repeat suicide attempts in patients 36 admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of the post-crisis suicide prevention long-term follow?up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts {Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). 37 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world. Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide (Pitman, Osborn, King, EL Erla ngsen, 2014}. Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion {Etzersdorfer 3t Sonneck, 1998; Niederkrotenthaler E: Sonneck, 2007}. While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care forthe bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions} and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. I Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief (Szumilas Kutcher, 2011}. 0 Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline}, and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood ofsuicide contagion. Potential Outcomes I Reductions in ideation/attempts 38 I Reductions in distress I Increases in treatment seeking - Improvements in reporting following suicide 0 Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others' suicide to reduce their own risk of suicide. One example of a postvention program, Stono'By Response Service (StondBy), provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs {Vissen Comans, Scuffham, 2014). In a study by Visser et al. (2014], StondBy clients were significantly less likely to be at high risk for suicidaiity (suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program {43% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention {versus passive approaches where survivors self-refer for services} (Cerel 8; Campbell, 2008). Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendetionsfor Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually {Niederkrotenthaler 8a. Sonneck, 200?}. Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition} have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010). Reports of individual suicidal ideation not accompanied by reports ofsuicide or suicide attempts, along with reports describing a "mastery" of a crisis situation where adversities were overcome, was 39 associated with significant decreases in suicide rates in the time period immediately following such reports (Niederkrotenthaler et al., 2010). Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a Convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector (including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Care. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront ofmany community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving 40 Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People At?Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 41 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical and/or medicolegal standards (Crosby, Ortega, et al., 2011}. uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems Crosby, Ortega, et al., 2011i Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of?death data from death certificates. is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2015]. Data from state and local Child Death Review teams and Suicide Death Review Teams (which are in a few states] offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic lnj'ury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc.], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department}. in addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade 42 students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The l?llotiomtzlI Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non?medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groUps at increased risk. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies (before risk occurs) and community?level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, 43 mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 44 References Archer, J., Bower, P., Gilbody, S., Lovell, K., Richards, 0., Gask, L., . . . Coventry, P. (2012). Collaborative care for depression and anxiety problems. Cochrane Database Rev, 10 (CD006525). ArsenauIt-Lapierre, (3., Kim, C., St Turecki, G. (2004). diagnoses in 3225 suicides: a meta-analysis. BMC 4, 37. K. E., Canham-Chervak, M., Black, S. A., Dada, E. 0., Millikan, A. M., Si. Jones, E. H. (2012). Mental health risk factors for suicides in the US Army, 2007?8. injury Prevention, 18(6), 405-412. Ba ndura, A. (1986). Sociolfounciations of thought and action: A social cognitive theory: Prentice-Hall, Inc. Barker, E., K?lves, K., De Leo, D. (2014). Management of Suicidal and Self-Harming Behaviors in Prisons: Systematic Literature Review of Evidence-Based Activities. Archives of Suicide Research, 18(3), 227-240. Beautrais, A. L. (2001). Effectiveness of barriers at suicide jumping sites: A case study. Aust NZJ 35. Beautrais, A. L., Gibb, S. J., Fergusson, D. M., Horwood, L. J., 8i Larkin, G. L. (2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukol, B. (2014}. Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the World Health Organization, 92(9), 641-655. Beyer, K. M., Layde, P. M., Hamberger, L. K., Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21(1), 49-54. doi:10.1127f1077801214564075 Blair, J. M., Fowler, K. A., Jack, 5. P., 8: Crosby, A. E. (2016). The National Violent Death Reporting System: overview and future directions. in} Prev, 22 Suppll, i?-ill. Bossarte, R. M, Karras, E., Lu, N., Tu, X., Stephens, B., Draper, J., 8i Kemp, J. E. (2014). Associations between the Department of Veterans Affairs' suicide prevention campaign and calls to related crisis lines. Public Heaith Rep, 129(6), 516?525. Branas, C. C., Cheney. R. A., MacDonald, J. M., Tam, V. W., Jackson. T. El. Ten Have, T. R. (2011). A difference? in-differences analysis of health, safety, and greening vacant urban space. Am Epidemioi, 1 74(11), 1296-1305. doi:10.1093iajeikwr273 Brown, G. K., Ten Have, T., Henriques. G. R., Xie, S. X., Hollander, J. E., 8; Beck, A. T. (2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5), 563?510. dni:1o.1001ijama.294.5.563 Bruce, M. L, Ten Have, T. R., Reynolds C. F., Katz, I. I., Schulberg, H. C., Mulsant, B. H., . . .Alextipoulos, G. S. (2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291(9), 1081?1091. Capaldi, D. M., Knoble, N. B., Shortt, J. W., 8; Kim, H. K. (2012). A Systematic Review of Risk Factors for Intimate Partner Violence. PartnerAbuse, 231?280. Carter, M, McGee, R., Taylor, 3., E: Williams, S. (2007). Health outcomes in adolescence: associations with family, friends and school engagement. 1 Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009). Strategic direction for the prevention of suicidal behavior: Promoting individual, family, and community connectedness to prevent suicidal behavior. Retrieved from: Strategic Direction Full version?apdf Centers for Disease Control and Prevention. (2016). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved December 23, 2016 Available at: 45 Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. doi:10.152llsuli.2008.38.1.30 Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Public Health Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. doi:10.152 llsuli.2007.37.5.543 Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Alcoholism: Clinical and Experimental Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Health Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. doi:10.1002lda.20895 Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health, 13(1), 1?12. Crosby, A. E., Espitia-Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. In Alcohol: Science, Policyr and Public Health. Oxford University Press. 190?193. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 55-13), 1-22. Crosby, A. E., Han, B., Ortega, L. A, Parks, S. E., 8.1 Gfroerer, J. (2011). Suicidal thoughts and behaviors among adults aged 24:18 years--United States, 2003-2009. MMWR Surveill Summ, 60(13), 1-22. Crosby, A. E., Ortega, L., 8; Melanson, C. (2011}. Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Crosby, A. E., 3: Sacks, J. J. (2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and life-threatening behavior, 32(3), 321-328. Cunningham, P. J. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Health Aff {Millwood}, 28(3), w490-501. doi:10.1327ihlthaff.28.3.w490 Curtin, S. C., Warner, M., 8: Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. NCHS Data Brief, no 241. Hyattsville, MD: National Center for Health Statistics. 46 Cylus, J., Glymour, M. M., 8: Avendano, M. (2014}. Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1968-2008. Am Epidemiol, 180(1), 45-52. 106 Dahlberg, L. L., Krug, E. G. {2002}. Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, R. Lozano World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. David-Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., Iskander, J. (2016}. CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR Morb Mortal Wkly Rep, 65, 894-987. Deisenhammer, E. A., lng, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009}. The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., Wintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 84 Levy, S. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., El. Giles, W. H. (2001}. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., Dunne?Maxim, K. (Eds). (1982). Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897l1951). Suicide: A study in sociology (JA Spaulding G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 189?). Edwards, R. W., Jumper?Thurma n, P., Plested, B. A., Oetting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 291?307. Escobedo, L. (3., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev, 34(5), 6894594. Etzersdorfer, E., 8; Sonneck, G. (1998). Preventing Suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 67-74. Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many ofthe leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245-258. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2}, 195?209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008}. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 703?209. doi:50042- 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., St Zhou, C. (2015). Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 64(38), 1074-1077. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 81. Frazier, L. (2015}. Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am 1 Public Health, 105(2), 311-316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. 47 Gilbody, S., Bower, P., Fletcher, J., Richards, 0., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternai medicine, 166(21), 2314-2321. Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. (2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. doi:10.1111isltb.12049 Gould, M. 5., Kalafat, J., Harrismunfakh, J. L., Eu Kleinman, M. (2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 3213), 338-352. Grossman, D. C., Mueller, 6. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), TOT-714. doi:10.1001ljama.293.6.707 Gruenewald, P. J., 8: Remer, L. {2006). Changes in outlet densities affect violence rates. Aicohoi Ciin Exp Res, 30(7), 1184-1193. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heaith Reports, 95(5), 411-421. Haegerich, T. M, 81 Dahlberg, L. L. (2011). Violence as a public health risk. American Joarnai oftifestyie Medicine, 392?406. 1559822611409127. Hailey, D., Raine, R., at Ohinmaa, A. (2008). The effectiveness of telemental health applications: a review. The Canadian Journai of 53(11), Hamby, 5., 81 J. (2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociaiagy. New York, NY: Springer. Han, B., Crosby, A. E., Ortega, L. A., Parks, 5. E., Compton, W. M., 84 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64 years in the United States. Comprehensive 66, 176-186. Harris, E. C., 8: Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta?analysis. 170, 205-223. Harris, E. C., St Barraclough, B. (1993). Excess mortality of mental disorder. Eir.I 173, 11-53. Harris, K. M., Carpenter, C., at Bao, Y. (2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian?Moghaddam, H., Sarjami, S., Kolahi, A. A., Carter, G. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital?treated self-poisoning. British Jaurnai of 198(4), 309316. Hawkins, J. D., Catalano, R. F., Kuklinski, M. R. (2014). Communities that ca re Encyciopedia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28(suppl 1), 4-9. Hayes, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 25(4), 431-456. Haves, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int;f Law 188-194. doi:10.1016ij.ijlp.2013.04.010 Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's Co-Occurring lnternalizing Q, 26(3), 189-201. Houle, J. N., Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am Pubiic Heaith, 104(6), 1073-1079. 48 Hunkeler, E. M., Katon, W., Tang, L., Williams, J. Kroenke, K., Lin, E. H., . . . Unutzer, J. (2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(7536), 259-263. Inagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad201412.048 lobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 640-653. 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8.: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12], 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. (2015). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 1-174. doi:10.15585immwr.ssE550631 Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, H., Giesbrecht, N., 8: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. S., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55?528. Kessler, R. (2., Demler, 0., Frank, R. (5., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. Engll Med, 352(24}, 2515?2523. doi:10.1056lN Elk/153043266 Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 223?279. doi:10.1027l0227? 591013000136 Klomek, A. E., Sourander, A., Gould, NI. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 8: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7423), 13?6. Knox, K. L., Pflanz, 5., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2457?2463. Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 2388. Leeb, R. T., Lewis, T., 81 Zolotor, A. J. (2011). A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of [lifestyle Medicine, 454-468. Lineberry, T. W., 8; O'Connor, 5. S. (2012). Suicide in the US Army. Mayo Clinic Proceedings, 82(9), 871-828. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. Z., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 757-766. Lipton, R., Gruenewald, P. (2002). The spatial dynamics of violence and alcohol outlets. .l Stud Alcohol, 63(2), 187-195. 49 Losel, F., 3L Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Maimon, D., Browning, C. R., 8; Brooks-Gunn, i. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. doi:10.1177l0022146510377878 Mark, T. L., Shern, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America?s mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., 8: Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis lntervention :9 Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8a Mortimer-Stephens, M. (2004]. Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8L Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243?253. Motto, J. A., 84 Bostrom, A. G. {2001). A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. {2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. Washington, D.C.: Author. National Action Alliance for Suicide Prevention Workplace Task Force. (2015). Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, D.C. Author. Niederkrotenthaler, T., at Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust 21 41(5), 419?428. doi:10.1080i00048670701265680 Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 Olfson, M., Gerhard, T., Huang, (2., 5., 81. Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. JAMA 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193-199. Perron, 5., Burrows, S., Fournier, M., Perron, P. A., 8: Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105lajph.2012.301089 50 Pirkis, J., Too, L. 5., Spittal, M. J., Krysinska, K., Robinson, J., St Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. The Lancet 2(11), 994- 1001. Pitman, A, Osborn, D., King, M., 8: Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. Plested, B. A., Edwards, R. W., 81. Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort Collins, CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Years intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rowhani-Rahba r, A, Simonetti, J. A, 8: Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemiol Rev, 38(1), 111-124. Ru benstein, L. V., Chaney, E. F., Ober, 5., Felker, B., Sherman, S. E., Lanto, A, at 'v'ivell, S. (2010}. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Families, Systems, ?5 Health, 28(2), 91-113. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A, Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. WestJ Emeryr Med, 17(1), 8-14. Rush, 3. R., Gliksma n, L., 3: Brook, R. (1986). Alcohol availability, alcohol consumption and alcohol?related damage. I. The distribution of consumption model. lStud Alcohol, 47(1), 1?10. Russell, 5. T., 8t Joyner, K. (2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Journal of Public Health, 91(8), 1276?1281. Shepard, D. 5., Gurewich, 0., Lwin, A. K., Reed, 6. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Lh?e Threat Behov, 46(3), 352?362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and ih?e? threatening behavior, 25(1), 10-21. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. EL, Kresnow, M. J., E: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M., Day, 5. X. (2002). Universal family-focused interventions in alcoholsuse disorder prevention: cost-effectiveness and cost-benefit analyses of two interventions. Stud Alcohol, 63(2), 219? 228. Stack, 5., 8t Wasserman, I. Economic strain and suicide risk: a qualitative analysis. Suicide and life? threatening behavior, 37(1), 103412. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. R., 8: Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Correct Health Care, 21(3), 222u242. doi:10.1127i1023345315587001 Sudak, H., Maxim, K., Carpenter, M. {2008). Suicide and stigma: a review of the literature and personal reflections. Acad 32(2), 136-142. Szumilas, M., 8; Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can Public Health, 102(1), 18-29. Toosi, M. (2015). Laborforce projections to 2024: The laborforce is growing, but slowly. Washington, D.C.: Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, (3. M., 8: Crawford, M. J. {2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-726. 51 US. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2015. Retrieved from US. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Ciinician Retention: A Story of Dedication and Commitment. Retrieved from US. Office of the Surgeon General, 81'. National Action Alliance for Suicide Prevention. (2012). 2012 Nationai Strategyfor Suicide Prevention: Goais and Objectives for Action. Washington, D.C.: HHS. US. Public Health Service. (1999). The Surgeon Generai's Caii to Action to Prevent Suicide. Washington, DC. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. [2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journaiof Community 42(1), 19-28. doi:10.1002fjcop.21586 Walrath, C., Ga rraza, L. G., Reid, H., Goldston, D. B., 8: McKeon, R. (2015). Impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am .i Pubiic Heaith, 105(5), 986-993. Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 106420. doi:10.1111isltb.12177 Wasserman, 0., Haven, C. W., Wasserman, (2., Wall, M., Eisenberg, R., Hadlaczky, (3., . . . Carli, (2014). School? based suicide prevention programmes: The SEYLE cluster?randomised, controlled trial. The Lancet, 385(9977), 15364544. Webster?Stratton, (2., Hammond, M. (1997). Treating children with early?onset conduct problems: a comparison of child and parent training interventions. Consuit Ciin 65(1), 93?109. Webster?Stratton, (2., Reid, M. J., 8: Hammond, M. (2001). Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Ciin Chiid 30(3), 283602. C. H., Reid, M. J., Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Joarnai of Ciinicai Chiid ii Adoiescent 40(2), 191?203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible years teacher and child training programs in high?risk schools. Journai of chiid and 49(5), 471-488. Wilcox, H. C., Kellam, S. G., Brown, C. H., Poduska, J. M., Ialongo, N. S., Wang, W., 8: Anthony, J. C. (2003). The impact of two universal randomized first- and secondegrade classroom interventions on young adult suicide ideation and attempts. Drug Aicohoi Depend, 95 Sappi 1, 560?73. Wilkins, N., Tsao, B., Hertz, M., Davis, R., Klevens, J. (2014). Connecting the Dots: An Overview of the Links Among Multiple Forms of Violence. Atlanta, GA: National Center for injury Prevention and Control, Centers for Disease Control and Prevention Oakland, CA: Prevention Institute. World Health Organization. (2013). Giobai and Regionai Estimates of Vioience Against Women: Prevaience and Heaith Effects of intimate Partner 1v?ioience and Non-Partner Sexuai Iv?ioience. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Giobai imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurray, M., Schmeelk-Cone, K., PetrovaWang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Pubiic Heaith, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., Chen, Y.-Y. (2012). Means restriction for suicide prevention. Lancet, 379(9834), 2393-2399. 52 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strategy Approach/Program, Practice or Policy Strengthen household financial security Suicide Best Available Evidence Suicide Attempts or ldeation Other Risk] Protective Factors for Suicide Lead Sectors1 Government (local, state, Federal) Access to Lethal Means (ED CALM) Unemployment benefit programs 1/ Strengthen - Business labor economic Other income supports 1" supports Housing stabilization policies Government (local, The National Neighborhood Stabilization v, state, Federal} Program Coverage of mental health conditions in health insurance policies Mental Health Parity Laws 1/ 1/ Strengthen access and Reduce provider shortages in underserved areas Government (local, state, Federal) 0? National Health Service Corps (NHSC) suicide care Healthcare Telemental health (TMH) Social services Safer surclde care through systems change Henry Ford Perfect Depression Care {Pre- v, V, cursor to Zero Suicide) Reduce access to lethal means among persons at-rislt Government local, intervening at suicide hot spots state) Create - Safe storage practices i/ - protective Publlc Health Emergency Department Counseling on v, Health ca re 50 Organizational policies and culture Best Available Evidence Together for Life US Air Force Suicide Prevention Program I Correctional suicide prevention v" BusinessiLabor Jus?ce Government (local, state, Federal) Community-based policies to reduce excessive alcohol use Government {locaL state) Promote connectedness Alcohol outlet density Business/labor Peer norm ro rams 3 Public Health Sources ofStrength Education Communitv engagement activities Public Health Greening vacant urban spaces Government (local, state] Teach coping and problem- solving skills Social emotional learning programs Public Health Youth Aware of Mental Health Program Education Good Behavior Game Parenting skill and family relationship approaches Public Health The incredible Years Education 51 Strengthening 10?14 Best Available Evidence Identify and at-risk Gatekeeper training support people Public Health Applied Suicide intervention Training Healthcare Garret Lee Smith Federai Grant Program 1/ v? Crisis Intervention Public Health . . . . . . Social Services National Sammie Prevention Lifeline 1/ Treatment for people at risk of suicide improving Mood Promoting Access to f, Caiiabarative Treatment Coiiaborative Assessment and Management Health ca re ofSuicia?aiity (CAMS) Diaiecticai Behavioral Therapy (DST) it? one ervlces Justice Attachment-Based Family Therapy (ABFT) v? Translating initiatives for Depression into E??ective Solutions project Treatment to prevent re~attempts ED Brief intervention with Follow-up Visits Health ca re Active foiiaw?up contact approaches v" Social services CBTfor Suicide Prevention 1/ 52 Best Available Evidence Postvention Lessen harms StondBv Response Service 1/ 1/ ea care and prevent future risk Safe reporting following a suicide Public Health 1/ Media Gurdelmes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. DRAFT NARRATIVE FOR CDC CLEARANCE The information in this narrative has not been cleared for dissemination. Please do not cite, copy or disseminate Preventing Suicide: A Technical Package of Policy, Programs, and Practices Prepared by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention 2017 Preventing Suicide: A Technical Package of Policies, Programs, and Practices is a publication of the National Center for Injury Prevention and Control of the Centers for Disease Control and Prevention. Centers for Disease Control and Prevention Thomas R. Frieden, MD, MPH, Director National Center for Injury Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested Citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. Preventing Suicide: A Technical Package of Policies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 2017. Contents Acknowledgements 4 External Reviewers 5 Overview of Technical Package 6 Strengthen Economic Supportsl4 Strengthen Access and Delivery of Suicide Care 17 Create Protective Environments 21 Promote Connectedness 26 Teach Coping and Problem-Solving Skills 28 Identify and Support People At-Risk 31 Lessen Harms and Prevent Future Risk 37 Sector involvement 39 Monitoring and Evaluation 41 Conclusion 42 References 44 Appendix A 50 Acknowledgements We would like to thank the following individuals who contributed in specific ways to the development ofthis technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but definitely not least, we extend our thanks and gratitude to all the external reviewers for their helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Office for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindrnan Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, lnc. Christine Schuler National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, lnc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Dan Reidenberg Suicide Awareness Voices for Education (SAVE) The experts above are listed with their affiliations at the time this document was reviewed. Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at-risk; and lessen harms and prevent future risk. The strategies represented in this package include those with a focus on preventing suicide from happening in the first place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy for Suicide Prevention and the National Action Alliance for Suicide Prevention?s priority to strengthen community-based prevention. Commitment, cooperation, and leadership from numerous sectors, including public health, education, justice, health care, social services, business, labor, and government support the implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a specific risk factor or outcome {Friedem 2014). Technical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The first component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the specific ways to advance the strategy. This can be accomplished through programs, policies, and practices. The approaches included come primarily from studies based in the United States. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision-making in communities and states. Preventing Suicide is a Priority Suicide, as defined by the Centers for Disease Control and Prevention (CDC), is part of a broader class of behavior called seif?o'irecteo? vioience. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potentiai for injury (A.E. Crosby, Ortega, Melanson, 2011). Self- directed violence may be suicidal or non-suicide! in nature. For the purposes of this document, we refer only to behavior where suicide is intended: I Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. attempt is defined as a non-fotolself?directed and potentially injurious behavior WIth any . . . Intent to die as a result of the behavior. A suicide attempt may or may not result In Injury Suicide is highly prevalent. Suicide presents a major challenge to public health In the UnIted States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs Office of the Surgeon General National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). In 2014 {the most recent year of available death data}, suicIde was responsible for 42,223 deaths in the which is approximately one suicide every 12 minutes (Centers for Disease Control and Prevention 2016}. In 2014, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes of death since 1925 in the 0.5 [Centers for Disease Control and Prevention, 2016}. lOverall suicide rates increased 24% from 1999 to 201:! [Curtin, Warner, SI Hedegaard, 2016]. Suicide is a problem throughout the life span; it Is the second leading cause ofdeath among those aged 10?34 years islethe fourth leading cause among persons in their 405, and seventh leading?au?e among persons in their 505. Suicide rates vary by racex?ethnicity, age, and other population characteristics, with the highest rates across the lifespan occurring among nonuHispanic American Indiaanlaska Native and nona Hispanic White population groups. In 2014, the rates for these groups were 1?.3 and 15 4 per 100,000 population, respectively {Centers for Disease Control and Prevention, 2016]. Other population groups disproportionately impacted by suicide include middle-aged adults {whose rates increased 43% from 1999 to 2014, with steep increases seen among both males and females aged 45?64 years; {CurtIn et al. ,2;016) Veterans and other military personnel [whose suicide rate nearly doubled from 2003 1 to 2008, surpassing the rate of suicide among civilians for the first time in decades? et al 2012; Lineberry 8; Connor}; bworkers in certain ochIpational groups le. protective service:r occupations; workers in farming, fishing, and forestry; McIntosh et al, 2016}; and lesbian, gay, bisexual, and/or queer (LGBOI youth, who experience increased suicidal ideatlon and behavior compared to their heterosexual counterparts ?(arm at al., 2016; Russell 8: Joyner, 2001) Suicides reflect only a portion of the problem Crosby, Han, Ortega, Parks, SI Gfroerer 2011) Substantially more people are hospitalized as a result of nonfatal suicidal behavior (Le suicide attempts} than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments} or not treated at all (HE. Crosby, Han, et al., 2011}. For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self-harm injuries, 27 who reported making a suicide attempt, and over 227 who reported seriously considering suicide ideationi (Ferdon et al In press} 15 associated with several risk and protective factors. Suicide, like other human behavlors has . . i no single determIning cause. Instead, suicide occurs in response to multiple biologlcal ogIcal Interpersonal environmental and societal influences that interact with one another often over time. Comment IAI: EndNote issue in formatting. Missing publication year for 'Lineberry BI O?Connor citation. The social-ecological model encompassing multiple levels of focus from the individual, relationship, community, and societal is a useful framework for viewing and understanding suicide risk factors identified in the literature (Dahlberg 8; Krug, 2002}. Risk and protective factors for suicide exist at each level. For example, risk factors include: Individual level: history ofdepression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants Relationship level: high conflict or violent relationships, sense of isolation and lack of social support, family/loved one?s history of suicide, financial and work stress 0 Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications) Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness (U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or whehave other noted risk factorsveoted, do not die by suicide. Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status (US. Office of the Surgeon General St National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014}. Protective factors, or those influences that buffer against the risk for suicide, can also be found across the different levels of the social-ecological model. Protective factors identified in the literature include: effective coping and problem solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means (US. Office of the Surgeon General 81 National Action Alliance for Suicide Prevention, 2012; World Health Organization, 2014). These protective factors can either counter a specific risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence le.g., child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increaseds-t?he risk of depression, post-traumatic stress disorder anxiety, suicide, and suicide attempts (Bossarte et al., 2014; D. P. Chapman et al., 2004; Dube et al., 2001; Felitti et al., 1998; Klomek, Sourander, 8t Gould, 2010; Leeb, Lewis, 8: Zolotor, 2011; World Health Organization, 2013}. Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence (WHO, 2013). Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse 8 problems and other instability, is associated with increasedsl-the risk for suicide and suicide attempts several fold {Bellis et al., 2014; Dube et al., 2001}. The effects of violence in childhood and adolescence can be observed decades later, including severe problems with finances, family, jobs, and stress factors that ?increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove beneficial in preventing suicide [Haegerich Dahlberg, 2011; Hamby 8i. 2013; Wilkins, Tsao, Hertz, Davis, 3; Klevens, 2014}. Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community lKIeiman, Riskind, Schaefer, 8t Weingarden, 2012}, school (Carter, McGee, Taylor, 8; Williams, 200?}, family {Maimon, Browning, E: Brooks-Gunn, 2010}, caring adults {Capaldi, Knoble, Shortt, 34 Kim, 2012; Losel 3f. Farrington, 2012}, and pro?social peers (Wyman et al., 2010} ?enhances resilience and help reduce risk for eta?suicide and other forms of violence. The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals, families, and communities (Dunne, McIntosh, El. Dunne- Maxim, 198?; Mishara, 1995; National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force, 2014; National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force, 2015}. in an early study, Crosby and Sacks {2002} estimated that 2% of the 0.5. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide. They also estimated that for each suicide, 425 adults were exposed, or knew about the death. In a more recent study, in one state, Cerel et al (2015} found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide andi'or having lived experience He, personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long-term health and mental health consequences ranging from anger, guilt, and physical impairment, depending on the means and severity of the attempt in. L. Chapman Si Dixon?Gordon, 2007}. Similarly, survivors of a loved one?s suicide may experience ongoing pain and suffering including complicated grief (Mitchell, Kim, Prigerson, E: Mortimer?Stephens, 2004}, stigma, depression, anxiety, post?traumatic stress disorder, and increased risk of suicidal ideation and suicide (Cerel, McIntosh, Neimeyer, Maple, 8: Marshall, 2014,- Sudak, Maxim, EL Carpenter, 2008}. Less discussed but no less important, are the financial and occupational effects for those left behind (Florence, Simon, Haegeric?n, Luo, Zhou, 2015}. The economic toll of suicide is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work?loss costs alone [Florence et al., 2015}. Adjusting for potential under-reporting of suicide and drawing upon health expenditures per capita, GDP per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self-directed violence to be approximately $93.5 billion in 2013 (Shepard, Gurewich, Lwin, Reed, 8: Silverman, 2015}. The overwhelming burden of these costs were from lost productivity over the life course, with the 9 i Comment Edits requested to avoid implying a causal link. i average cost per suicide being over $1.3 million {Shepard et al., 2016). The true economic costs are likely higher, as neither study included monetary figures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable (US. Public Health Service, 1999). While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or approach (Silverman E: Maris, 1995; U.S. Office of the Surgeon General 8: National Action Alliance for Suicide Prevention, 2012}. Rather, suicide prevention is best achieved by a focus across the individual, relationship, family, community, and societal-levels and across all sectors, private and public le.g., business, public health, physical and behavioral healthcare, justice, education, and labor; National Action Alliance for Suicide Prevention, 2014; World Health Organization, 2014L Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a} meta-analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design} evaluation study that found significant preventive effects on suicide; c) meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d} evidence from at least one rigorous RCT or quasi-experimental design) evaluation study that found significant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving beneficial effects on multiple forms of violence; no evidence of harmful effects on specific outcomes or with particular subgroups; and feasibility of implementation in a U.S. context if the program, policy, or practice has been evaluated in another country. Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide help-seeking, stigma reduction, depression, connectedness). In terms of the strength of the evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors reflect the developing nature of the evidence base and the use of the best available evidence at a given time. it is also important to note that there is often significant heterogeneity among the programs, policies, or practices that fall within one approach or strategy area in terms of the nature and quality of the available 10 evidence. Not all programs, policies, or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations. Tailoring programs and more evaluation may be necessary to address different population groups. The evidence-based programs, practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have beneficial effects on risk or protective factors for suicide. Context and Cross-Cu tting Themes One important feature of the package is the complementary, but potentially synergistic impact of the strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box below). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included first, followed by those that might impact more select populations persons who have already made a suicide attempt). Preventing Suicide Strategy Approach Strengthen economic supports I Strengthen household financial security I Housing stabilization policies Strengthen access and delivery of suicide I Coverage of mental health conditions in health care insurance policies Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments Reduce access to lethal means among persons at? risk of suicide I Organizational policies and culture I Community-based policies to reduce excessive alcohol use Promote connectedness I Peer norm programs I Community engagement activities Teach coping and problem-solving skills I Social?emotional learning programs I Parenting skill and family relationship approaches Identify and support people at risk I Gatekeeper training I Crisis Intervention I Treatment for people at risk of suicide Treatment to prevent re-attempts Postvention I Safe reporting following a suicide Lessen harms and prevent future risk 11 It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social emotional learning programs, an approach under the Teach Coping and Problem- Soiving strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative influences substance use) associated with suicide. The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden of suicide. Suicide ideation, thoughts, attempts, and deaths vary by gender, race/ethnicity, age, occupation, and other important population characteristics. Further, certain transition periods are also associated with higher rates of suicide transition from working into retirement, transition from active duty military status to civilian status). In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced {Turecki, 2014). Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience. Identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or beneficial effects on risk or protective factors for suicide is only the first step. In practice, the effectiveness ofthe programs, policies and practices identified in this package will be strongly dependent on how well programs are implemented, as well as the partners and communities in which they are implemented. Practitioners in the field may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work see Edwards, Jumper-Thurman, Plested, Oetting, Swanson, 2000; Hawkins, Catalano, 3; Kuklinski, 2014; Plested, Edwards, 8: Jumper-Thurman, 2006). These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s profile of risk and protective factors with evidence-based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program 12 materials, training and technical assistance] can also influence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re-attempts}. The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sector involvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. 13 Strengthen Economic Supports Rationale Studies from the U.S. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working- age individuals 25 to 64 years old {Luo et al., 2011; Fowler et al., 2015). Economic and financial strain, such as job loss, long periods of unemployment, reduced income, difficulty covering medical, food, and housing expenses, and even the anticipation of such financial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems. Buffering these risks can, therefore, potentially protect against suicide (Stack 8a Wasserman, 2007). For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household financial security and ensuring stability in housing during periods of economic stress. 0 Strengthening household financial security can potentially buffer the risk ofsuicide by providing individuals with the financial means to lessen the stress and hardship associated with a job loss or other unanticipated financial problems. The provision of unemployment benefits and other forms of temporary assistance, livable wages, medical benefits, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household financial security. I Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of financial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modification programs, move-out planning, or financial counseling services that help minimize the risk or impact of foreclosures and eviction. 14 Potential Outcomes [Comment IAI: is this intentionally vague? Why not just state TANF. I Reductions in foreclosure rates I Reductions in eviction rates I Reductions in emotional distress I Reductions in suicide Evidence There is evidence suggesting that strengthening household financial security and stabilizing housing can reduce suicide risk. 0 Strengthen household financial security. The Federal-Store Unemployment insurance Program allows states to define the maximum amount and duration of unemployment benefits that workers are entitled to receive after a job loss {Cylus, Glymour, SI Avendano, 2014]. An examination of variations in unempioyment bene?t programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment benefits (mean level: 5?,990 per person in U.5. constant dollars; Cylus et al., 2014]. The effects of unemployment benefit programs were also consistent by sex and age group. Another US. study examining the iink between unemployment and suicide rates using suicide dataI length of unemployment (less than 5 weeks, 5-14 weeks, 15-25 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of job, predicted suicide risk {Classen 8t Dunn, 2012). Together, these results suggest that not only should state unemployment benefit programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen household financial security le.g., transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical benefits, and other forms of family assistance) have also shown an impact on rates of suicide. A study by Flavin and Radcliff (2009} examined the impact of states? per capita spending on transfer payments, medical benefits, and family assistance (Temporary Assistance to Needy Families and total state spending on suicide rates between 1990?2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical benefits, and family assistance increased there was an associated decrease in state suicide in-general matwas-asseeatedwith-the seductionbeespending terms of lives saved, Flavin St Radcliff calculated the cost of reducing a state?s suicide rate by a full point for the years studied. At the national level, they estimated that-3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of assistance by $45 per year (Flavin 84 Radcliff, 2009]. Although this was a correiational studythe results demonstrate the potential benefits of policies that reach particularly vulnerable individuals during periods of great need and increased risk for suicide. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Nationai Neighborhood Stabilization Program was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration ofthe neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuvers. This program also offers financial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorouslv evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analvsis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults (Houle Light, 2014}. Another studv of data from 16 U.S. states participating in the National Violent Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began} to 2010 (after it had peaked; Fowler, Gladden, Vagi, Barnes, and Frazier (2015)). Most of these suicides occurred prior to the actual loss of the decedent's home. These findings suggest that integrating suicide prevention resources, messaging, and referrals into financial, foreclosure, and move-out planning and counseling services may help to prevent suicide. 16 Strengthen Access and Delivery ofSuicide Care Rationale While most people with mental health problems do not attempt or die by suicide (Olfson, Gerhard, Huang, 8: Stroup, 2015; Owens, 2002), and the level of risk conferred by different types of mental illness varies {Arsenault?Lapierre, Kim, El Turecki, 200d; E. C. Harris 8: Barraclough, 1997; Tyrer, Reed, 81 Crawford, 2015}, previous research indicates that mental illness is an important risk factor for suicide (E. C. Harris 81 Barraclough, 15198; World Health Organization, 2014). State?level suicide rates have also been found to be correlated with general mental health measures such as depression [Lang 2013; Mark, Shem, Bagalman, St Cao, 200?}. Findings from the National Comorbidity Survey indicate that relatively few people in the US. with mental health disorders receive treatment for those conditions (Kessler et al., 2005}. Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services (Cunningham, 2009). Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention {World Health Organization, 2014). Additionally. research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and efficiently deliver such care Coffey, 200?]. Apart from treatment benefits, these approaches can also normalize help-seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: I I i Coverage of mental health conditions in health insurance policies. Federal state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns ii.e., mental health parity}. Benefits and services covered include such things as the number of visits, co?pays, deductibles, inpatientfoutpatient services, prescription drugs, and hospitalizations. If a state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. If a state has a weaker parity law than the federal parity law includes coverage for some mental health conditions but not others}, then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which benefits and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. I Reduce provider shortages in underserved areas. Access to effective and state?of?the?art mental health care is largely dependent upon the training and the size ofthe mental health care workforce. Over 35 million Americans live in areas with an insufficient number of mental health 1? Comment IAI: Double check and make sure that this will not need an ACA review providers; this shortage is particularly severe among low?income urban and rural communities (US. Department of Health and Human Services Health Resources and Services Administrations, 20153}. There are a member-of ways to increase the number and distribution of practicing mental health providers in underserved areas including offering financial incentives through existing state and federal programs loan repayment programs} and expandingtelemental y" health services. Such approaches can increase the likelihood that those in need will be able to 3? access affordable, quality care for mental health problems, which can reduce risk for suicide. Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this is just one piece of the puzzle. Care should also be deiivered efficiently and effectively. More specifically, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identification and assessment of suicide risk, implementation of evidence-based treatments (see identr? and Support People Alt-Risk, p.31], continuity of care, and continuous quality improvement. Care that is patient?centered and promotes equity for all patients is also of critical importance [National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force, 2014). Potential Outcomes - Increases in access to mental health services 0 Increase in utilization of mental health services Reductions in of mental illnesses and suicidality - Reductions in rates of suicide attempts I Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can red Lice risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey of Drug Use and Health is a nationally representative survey of the U.S. population that provides data on substance use, mental health conditions, and services utilization. Using data from this survey, K. M. Harris, Carpenter, and Bao (2006) found that 12 months after states enacted mentai heoith parity tows, self-reported use of mental healthcare services significantly increased. Moreover, subsequent research by Lang (2013] examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, specifically, were associated 18 9 I I I .?(C?mment define. with an approximate 5% reduction in suicide rates. This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year (Lang, 2013). Reduce provider shortages in underserved areas. One example of a program to improve access to mental health care providers is the Notional Health Service Corps (NHSC), which offers financial incentives to attract mental/behavioral health clinicians to underserved areas. Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to The National Health Service Corps (U.S. Department of Health and Human Services Health Resources and Services Administrations, 2016b]. Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Telementol health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance. TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health ca re system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance abuse, and suicidal ideation and suicide deaths among other outcomes (Hailey, Roine, 8: Ohinmaa, 2008}. Further, Mohr and colleagues (2008} conducted a meta-analysis examining the effect of delivered specifically via telephone and found that it significantly reduced depressive in comparison to face?to-face They also found that treatment attrition rates were significantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy. Th us, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford ?healthcare which is a large health maintenance organization (HMO) in the state of Michigan, pioneered the Perfect Depression Core program, the pre-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve "breakthrough improvement" in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, efficiency, and equity among patients. The program screened and assessed each 19 patient for suicide risk and implemented coordinated continuous follow-up care system wide (C. E. Coffey, 2005). An examination of the impact of the program found that there was a dramatic and statistically significant decrease in the rate of suicide between the baseline years, 1999 and 2000, prior to the intervention to the intervention years, 2002-2009. During this time period, the suicide rate fell by 82% (C. E. Coffey, 2006; C. E. Coffey, Coffey, 81 Ahmedani, 2013). Further, among HMO members who received mental health specialty services, the suicide rate significantly decreased over time from 1999 to 2010 {110.3 to 47.6 per 100,000 population; p<.04) with a mean of 36.2 per 100,000 over the period. Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from to 5.6 per 100,000 Similarly, in the state of Michigan, rates ofsuicide in the general population increased over the period from 9.8 to 12.5 per 100,000 {p<.001) (M. Coffey, Coffey, 8; Ahmedani, 2015}. 20 Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes [Haddon, 1980). Creating environments that address risk and protective factors where individuals live, work, and play; can help prevent suicide (Dahlberg 8i Krug, 2002; U.S. Office of the Surgeon General 8; National Action Alliance for Suicide Prevention, 2012}. For example, rates of suicide are high among middle-aged adults {who comprise 42.6% of the workforce; Toosi, 2015}; among certain occupational groups farming, fishing, forestry, and construction; Han et al., 2016; McIntosh et al., 2016), and among people in detention facilities leg. jail, prison}, to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are not (Knox et al., 2010; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). Similarly, modifying the characteristics of the physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition {Beautrais, Gibb, Fergusson, Horwood, 8: Larkin, 2009; A.E. Crosby, Espitia-Hardeman, Ortega, 8t. Lozano, 2013; Kaplan et al., 2013; Miller, Warren, Hemenway, 8i Azrael, 2015; Runyan et al., 2016; Stokes, McCoy, Abram, Byck, Teplin, 2015}. Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. Reduce access to lethal means among persons at-risk of suicide. Means of suicide such as firearms, hanging/suffocation, orjumpingfrom heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a firearm in a suicide attempt will die from the injury). Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 minutes (Deisenhammer et al., 2009; Simon et al., 2001), and 2) thatpeople tend not to substitute a different method when a highly lethal method is unavailable or difficult to access (Hawton, 2007; Yip et al., 2012). Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more difficult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at-risk of suicide: 21 intervening otSuicide Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help {Cox et al., 2013). Safe Storage Practices. Safe storage of medications, firearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing firearms-docked in a secure place in a gun safe or lock box), unloaded and separate from the ammunition--and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior attempts {Rowhani-Rahbar, Simonetti, 8L Rivara, 2016; Runyan et al., 2016). 0 Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments leg. residential settings). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, financial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolation) (Hayes, 2013; National Action Alliance for Suicide Prevention Workplace Task Force, 2015). a Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol-involved suicides {Escobedo E: Ortiz, 2002; Giesbrecht et al., 2015). Policies to reduce excessive alcohol use broadly include zoning to limit alcohol outlet locations and density, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age. These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts (Cherpitel, Borges, Wilcox, 2004). Potential Outcomes Increases in safe storage of lethal means Reductions in rates of suicide Reductions in suicide attempts 22 I Reductions in suicide deaths I Increases in help-seeking Reductions in alcohol-related suicide deaths Eviden ce The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at-risk of suicide. A meta-analysis examining the impact of suicide hotspat interventions implemented in combination or in isolation, both in the U.S. and abroad, found associated reduced rates of suicide (Cox et al., 2013; Pirkis et al., 2015). For example, after erecting a barrier on the Jacques-Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year (Perron, Burrows, Fournier, Perron, Ouellet, 2013). Moreover, the reduction in suicides by jumping was sustained even when all bridges and nearby jumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites (Perron et al., 2013). Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased five; fold (Beautrais, 2001; Beautrais et al., 2009). Another form of means reduction involves implementation of safe storage practices. in a case- control study of firearm-related events identified from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, Grossman et al. (2005) found that storing firearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents. Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices significantly increased safe firearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one?s own (Rowhani-Rahbar et al., 2016). Another program, The Emergency Department Counseling on Access to Lethal Means (ED CALM), trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 18 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al. {2015) found that at post- test 76% (of the 55% of parents followed up, n=114) reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre?test (Le. all (100%) reported guns were currently locked up at post-test {Runyan et al., 2016). 23 Organizational policies and culture. Togetherfor Life is a workplace program of the Montreal Police Force implemented to address suicide among officers. Policv and program components were designed to foster an organizational culture that promoted mutual support and solidaritv among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifving suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking {Mishara St Martin, 2012}. Police suicides were tracked over 12 vears and compared to rates in the control city of Quebec. The suicide rate in the intervention group decreased significantly by 78.9% to a rate of 6.4 suicides per 100,0{10 population per vear compared to an 11% increase in the control citv {29.0 per 100,000; Mishara Martin, 2012). Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 11 policv and education initiatives and was designed to change the culture of the Air Force surrounding suicide. The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training), and investigates even.I suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solelv as medical problem and instead sees them as larger serviced wide problems impacting the whole community (Knox, Litts, Talcott, Feig, 84. Caine, 2003). Using a time?series design to examine the impact of the program on Various violence-related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide {Knox et al., 2003}.The program was also associated with relative risk reductions in related outcomes including moderate and severe famiiv violence {30% and 54%, respectivelv), homicide and accidental death {Knox et al., 2003}. A longitudinal assessment of the program over the period 1981 to 2008 [15 vears before the 199? launch of the program and 11 years post-launch) found significantiv lower rates of suicide after the program was launched than before launch (Knox et al., 2010). These effects were sustained over time, except in 20m,_ which the authors found was associated with less rigorous the other 1veers [Knox et al., 2010]. Finallv, while the evidence is still being built for suicide prevention in preliminary evidence suggests teat?organizational policies and practices that include routine suicide prevention training for all staff, standardized intake screening and risk assessment, provision of shared information between staff members, especially in transitioning or transferring of inmates, varving levels of observation, safe phvsical environment, emergencv response protocols, notification of suicidal behavior/suicide through the chain of command, and critical incident stress debriefing and death review can potentlallv reduce suicide. When these policies and practices were implemented across 11 state prisons in Louisiana, suicide rates 24 Comment IAI: What does this mean and pertinence to ?delity? [dropped from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the CommentlAl: Computea45%reduction following vearkHaves, 1995}. Other similar programs have seen declines in suicide both in the in the fatE- this large effect. United States and internationallv (Barker, K?lves, 8; De Leo, 2014]. 3 Comment IAI: Citations? f; Comment [Ah Please clarify the results I from this study, did this result in an I Community-based policies to reduce excessive alcohol use. While multiple policies to limit . i' increase in suicide or decrease? Not clear. excessive use of alcohol exist, Eeverai studies]on alcohol outlet density, specificallv, suggest that . measures to reduce alcohol outlet densitv can potentiailv reduce alcohol-involved suicides. 5' outlet densitv, suicide mortalitv, and ,5 Additionally, a longitudinal analvsis of alcohol hospitalizations for suicide attempts over 6 veers in 581 California zip codes indicated that the densityr of bars, specifically, is Enverseldrelated to suicide and suicide attempts, particularly in rural areas (Johnson, Gruenewald, Remer, 2009i. 25 Promote Connected ness Rationale Sociologist, Emile Durkheim theorized in 1397 that weak social bonds, i.e. lack of connectedness, are among the chief causes for suicidality (Durkheim, Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others (Centers for Disease Control and Prevention, 2009}. Social connections can be formed within and between multiple levelsofthe social ecology (Dahlberg 8t Krug, 2002}, for instance between individuals leg. peers, neighbors, co-workers), families, schools, neighborhoods, workplaceg, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations (Beyer, Layde, Hamberger, Ea. Laud, 2015; Muennig, Cohen, Palmer, 31 Zhu, 2013). Many ecological cross?sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest the?patteen?is?tewasds?a positive association between social capital measured by social trust, communityfneighborhood engagement, and improved mental health. Connectedness and social capital together ean-seeve tom protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, increasing belongingness, bersonal value,. and worthL all?ef?wleiehgg helps individualete-build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs of its members and provide collective primary prevention activities to the community as a whole (Centers for Disease Control and Prevention, 2009). Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. - Peer norm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults. and promote peer connectedness. By leveraging the Comment IAI: Add verb here to be consistent with sentence structure, suggest rewriting this sentence or re order? leadership qualities and social influence of peers, these approaches can be used to shift group- level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings. Community engagement activities. Community engagement is an aspect of social capital. I Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and physical exercise. These activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in 25 Comment IAI: Edit requested because enhanced overall phvsical health. reduced stress. and decreased depressive therebv this evidence is mixed. Their more recent results were not signi?cant. reducing risk of suicide. Potential Outcomes Reductions in maladaptive coping attitudes and behaviors Increases in healthv coping attitudes and behaviors Increases in referrals for vouth in distressed Increases help-seeking behaviors Increases in positive perceptions of adult support Evidence Current evidence suggests a number of positive bene?ts of peer norm and communitv engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high?schools (E metropolitan. 12 rural}, vaan at al. {2010} found that the program improved adaptive norms regarding suicide among peer leaders. connectedness to adults. and school engagement. Peer leaders Were also more likely than controls to refer a suicidal friend to an adult. For studEnts, the program resulted in increased perceptions of adult support for suicidal vouths, particularlv among those with a history of suicidal ideation, and the acceptability of help-seeking behaviors. Finallv, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders (vaan at al., 2010]. Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and communitv members. worked together to green 4,436 lots [or 7.8 million square feet} in 4 areas of the citv. Researchers found significant reductions in community residents? self?reported level of stress, which is a risk factor for suicide. and engagement in more phvsical exercise, a protective factor for suicide. than residents in control vacant lot areas. There is some evidence for ocher benefitsgincludi?ged reductions in firearm assaults and vandalism EBranas et al., 2011} 2? Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, conflict resolution, and critical thinking. Life skills are important in shielding individuals from suicidal behaviors (World Health Organization, 2014]. Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories (Bandura, 1986), surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness] characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and find solutions for problems has been characterized among suicide attempters {Pollock 8: Williams, 2004}. Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. . Social emotional learning programs focus on develooing and strengthening communication and problem-solving skills, emotion regulation, conflict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth to address other negative influences le.g., substance use} associated with suicide. These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work (Herman, Borden, Reinke, &Webster-Stratton, 2011). I Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities. Programs are typically designed for parents or caregivers with children in a specific age range and can be self-directed or delivered to individual families or groups of families. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Specific program content typically varies by the age of the child but often has consistent themes of child development, parent-child communication and relationships, and youth?s interpersonal and problem-solving skills. 28 Potential Outcomes Comment IAI: Explain the difference .- Reductions in suicide attempts and suicide ideation absolute and relative risk for lair - Reductions in suicide risk behaviors ii.e., depression, anxietv, conduct problems, substance abuse] readers; cum ?sport Mame mk' Improvements in help?seeking behavior Improvements in social competence and emotional regulation skills Improvements in problem-solving and conflict management skills Evidence Several social emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse (M. 5. Knox, Burkhart, 3t Hunter, 2010). a Social emotional learning programs. The Youth Aware of Mental Health Program (YAM) is a program developed for teenagers aged 14-16 that uses interactive dialogue and role-plaving to teach adolescents about the risk and protective factors associated with suicide {including knowledge about depression and anxiety} and enhances their problem-solving skills for dealing with adverse life events, stress, school and other problems (Wesserman et al., 2014). In a cluster-randomized controlled trial conducted across ID European Union countries and 168 schools, students in schools randomized to mm were significantlv less likelv to attempt suicide and have severe suicidal ideation at the 12-month follow-up compared to students in control schools which received eduoational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, five attempted suicide in the MM group compared to 11 in the control group. Additionallv, related to severe suicide ideation, in the YAM grouppbsolute risk fell by {1.50% and relative risk fell by 49.6% (Wesserman et al., 2014).] I 1 Another example is the Good Behavior Game {686), which is a classroom?based program for elementarv school children aged 6-10. The program uses a team-based behavior management strategv that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior [Wilcox et al., 2008}. Two cohorts of vouths participated in the program in 1985-86 and 1986?8? school vears when thev were in the first and second grades. A number of proximal and distal outcomes Were assessed among the two cohorts over time. With respect to distal suicide?related outcomes, an outcome evaluation of the EEG indicated that individuals in the first cohort who were assigned to participate in 636' when they Were in the first grade reported half the adjusted odds of suicidal ideation 29 and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting. The beneficial effect ofthe program was consistent for suicidal ideation regardless ofwhether baseline covariates were included. The EEG effect on attempts was less robust in some adjusted models including caregiver mental health. in the second cohort of EEG students, neither suicidal ideation nor lComment How so? I 1 i; Comment IAI: in the last version you i had added ?and pointed to the need for a. 636 to be delivered with precision, consistency, and teachersupport." It could help to put something like this back i' in at the end of this sentence. suicide attempts were significantly different between 636 and the control interventions 3. fl. {Comment One or both cohort? {Wilcox et al., 2008). [The researchers believed this may have been due to a lack of implementation Fidelity]. EEG was also found to be associated with Eeduced risk of later j[ substance aims? a risk factor for suicide lKellam et al., 2008). Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the lncredibie rears {it} is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems, two important suicide risk factors in youth by improving protective factors such as responsive and positive parent-teacher?child interactions and relationships, emotion self?regulation and social competence [all protEctive factors for suicide} (Herman et al., 2011]. The program includes 9?20 sessions offered in community?based settings religious, recreation f! Please con?rm. Comment IAI: Specific outcomes or all outcomes? Comment IAI: Number of sessions? Do Herman et al. indicate the ideal number of sessionsL?arlge from 9 to 20Er centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the 0" program on reducing internalizing such as anxiety and depression, and child conduct problems (C. H. Webster-Stratton, Reid, Beauchaine, 2011; Webster?Stratton, iamila Reid, 3: Stoolmiiler, 2008). The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1-year follow?up (Reid, WebstenStratton, Hammond, 2003,- C. Webster?Stratton Ea Hammond, 1997; C. Webster-Stratton, Reid, 8: Hammond, 2001). The program demonstrated greater fene?ts]as theEosageJof the intervention increased {Herman et al., 2011}. Additionally, Strengthening 10?14 is a program that involves sessions between parents, youth, and family with the goal of improving parents? skills for disciplining, managing emotions and conflict, and communicating with their children; promoting youths' interpersonal and problem~solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide i5poth, Guyll, Day, 2002}. Strengthening has been shown to significantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families [Spoth et al., 2002}. 30 Jk. Identify and Supportf?eople MrRisk] Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio?economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority stems; and members of certain racial and ethnic minority groups et al., 2012; Centers for Disease Control and Prevention, 2016; Curtin et al., 2015; Kann et al., 2016; Lineberry 8L O'Connor; Russell 8; Joyner, 2001}. Supporting these at-risk groups requires proactive case finding and effective response, crisis intervention, and evidence?based treatment. Finding optimal ways of identifying at-risk individuals, customizing services to make them more accessible linternet?based services when appropriate} and engaging people in evidence?based care through such measures as collaborative treatment], remain key challenges. Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in ?disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services. Approaches The following approaches focus on identifying and supporting people at increased risk. I Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk. I Crisis intervention. These approaches provide support and referral services, typicaliy by connecting a person in crisis (or a friend or family member of someone at~risk} to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feeiings of depression, hopelessness, and subsequent mental health care utilization. Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Treatment for people at-risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors 31 .- Suggest: At?Rlsk People 1 with problem-solving and emotion regulation. Treatment usually takes place in a onez-ongone or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative between patient and therapist or care manager) and/'or integrated care linkage between primary care and behavioral health care} can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk {Archer et al., 2012; Bruce et al., 2004; Gilbody, Bower, Fletcher, Richards, 8: Sutton, 2005}. Treatment to prevent re-attempts. These approaches typically include follow-up contact and use diverse modalities home visits, mail, telephone, e-mail) to engage recent suicide attempt survivors in continued treatment to prevent re-attempts. Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one- on-one interpersonal therapy and/or group therapy. Approaches that engage and connect attempters to peers and providers are especially important because many attempters do not present to aftercare; 1296?2594: reattempt within a year, and 396-996 of attempt survivors die by suicide within 1 to 5 years of their initial attempt (Inagaki et al., 2015) Potential Outcomes Reductions in suicide attempts Reductions in suicide deaths Reductions in of mental illnesses and suicidal ideation Reductions in mental health-related sequelae Reductions in re-attempts Increases in connectedness Improvements in coping skills Increases in identification of individuals at-risk for suicidal behavior Increases in treatment engagement by at-risk individuals Increases in community members trained to identify at-risk individuals Increases in referrals for health care Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision of treatment and support for these individuals can positively impact suicide and its associated risk factors. Gatekeeper training. Applied Suicide intervention Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to 32 identifv and connect with suicidal individuals, understand their reasoning for living and dving, and assist with safely connecting those in need to available resources. In a study emploving a i randomized controlled trial, Gould, Cross, Pisani, Munfakh, it Kleinman (2013) evaluated the i training across the Notionoi Suicide Prevention Lifeiine network of hotlines over the period 2003- 2009. iJsing data from 1,410 suicidal individuals who called 17 Lifeline centers, the researchers fl found that compared to callers who spoke to counselors that received the usuai care training, i individuals who spoke with counselors without training in were signi?cantly more likelv to i feel depressed, suicidal, more overwhelmed, and less hopeful by the end of their call to the trained in were also more i skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASIST did not result in more comprehensive suicide risk assessments than usual care training (Gould et al., 2013]. Comment IAI: Awkward sentence structure. Please revise. Gatekeeper training has also been a primaryr component of the Garret Lee Smith (GL5) Suicide Prevention Program, which is in place in 49 states and 48 tribes. A multi?site evaluation assessed the impact of communityr gatekeoper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among voung people aged 10-24 in counties implementing GLS trainings, with the trajectorv observed in similar counties that did not implement these trainings. Counties that implemented GL5 trainings had significantlv lower vouth suicide rates one year following the training implementation [Walrath, Garraza, Reid, Goldston, 8: McKeon, 2015]. This finding equates to a decrease of 1 suicide death per 100,000 among vouth ages 10 to 24, or the prevention of 23? deaths in the age group, between 2002ir and 2010. Counties impiementing (its program activities also had significantlv lower suicide attempt rates among vouth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GLS activities {4.9 fewer attempts per 1000 vouths; Godov Garraza, IUlialrath, Goldston, Reid, 8; McKeon, 2015]. More than 79,000 suicide attempts may have been prevented during the period-examined, following implementation of the 615 program. Crisis intervention. Suicide prevention hotlines are one waifr to provide crisis intervention. In an evaluation of the effectiveness of the Notionoi Suicide Prevention tifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow?up assessment between 1 and 52 davs [mean=13.5 davs) after the initial assessment. Researchers found that over half of the initial sample were seriouslv considering suicide when thev called, and thev had a plan for their suicide. Researchers also found that among follow?up participants, there was a significant decrease in pain, hopelessness, and intent to die between initiation of the call [time 1) to follow?up [time 3). Between time 2 {end of the call) to time 3, the effect remained for pain and 33 If hopelessness, but Was not signi?cant for intent to did LGould, Kalafat. Harrismunfakh. 8t! 200i} Kleinman, "i I Treatment for people at-risk of suicide. The improving Mood Promoting Access to Coiiaborative Treatment program aims to prevent suicide among older primarv care patients bv reducing suicide ideation and depression. facilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up (biweekl?vr during an acute phase and during continuation phase) by a depression care manager (Hunkeler et al., 2006). The program has been shown to significantlv improve qualitv of life, and to reduce functional impairment, depression and suicidal ideation over 24?months of follow-up [Hunkeler et al., 2006; Unutzer et al., 2006] relative to patients who I I I I I received care as usual. Collaborative Assessment and Management ofSaicia?aiity (CAMS), is a therapeutic approach for suicide?specific assessment and treatment. The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-specific treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS beenwitested and supported in E- correlational studies (Jobes, 2012). in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under wav. A feasibility trial with a community-based sample of suicidal outpatients randomlv assigned to CAMS or enhanced care as usual {intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed) found better treatment retention among the CAMS group and significant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow?up [Comtois et al., 2011). Other examples include Dialectico! Behavioral Therapy and Attachment?Based Family Therapy a multicomponent therapyr for individuals at high risk for suicide and who i mav struggle with impulsivitv and Eamotional regulatlorlir?l'he components of?EJB?T?includej individual therapy, group skills training, between-session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likelv to make a suicide attempt at two-vear follow?up than women receiving communitv treatment [23% vs required less hospitalization for suicide ideation, and had lovver medical risk across all suicide attempts and self?injurious acts combined [Linehan et al., 2005]. 34 I. I Comment IAI: This text is somewhat confusing. The fact that the effect for intent to die did not "remain? signi?cant suggest that it went back up to what it was when thev initiated the call. Also this ends on a disconcerting note. in the other places where you had an important null result you included an idea about why and what was needed leg, Can you i. Ldo that here too? [Comment Fix formatting Comment IM: Emotion regulation issues? Or emotion dvsregulationABFT is a program for adolescents aged 12-18 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety (Diamond et al., 2010). A randomized controlled trial of ABFT found that suicidal adolescents assigned to ABFT experienced significantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a significantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%) and at 24 weeks {82.1% vs. 46.2%) (Diamond et al., 2010). The Veterans Affairs Translating initiatives for Depression into Ejjfective Soiutions project uses a depression care liaison to link primary care and mental health services. The depression care liaison assesses and educates patients and follows up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the efficiency of providing mental health services by bringing mental health care to the primary care setting, where most patients are first detected and subsequently treated for many mental health conditions. An evaluation of found significant decreases in depression severity scores among 70% of primary care patients {Rubenstein et al., 2010}. also demonstrated 85% and 95% compliance with medication and follow-up visits, respectively (Rubenstein et al., 2010]. Treatment to prevent re-attempts. Several strategies that aim to prevent re?attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Foiiow-ap Visits is a program that involves a one-hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow-up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12,13 months}. Follow-up contacts are either conducted by phone or through home visits according to a specific time?line for up to 18_~months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in five countries (Brazil, India, Sri Lanka, Iran, and China) found that a brief intervention combined with Q-rnifollow-up visits over 18_?months was associated with significantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively} {Fleischmann et al., 2008). Another example of treatment to prevent re-attempts involves active foiiow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation. These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the first 12 months post- discharge with some programs continuing contact for two or more years}. In a meta-analysis conducted by Inagaki et al. {2015), interventions to prevent repeat suicide attempts in patients 35 admitted to an emergency department for suicide attempt were found to reduce reattempts by approximately 17% for up to 12 months post-discharge; however, the effects of these approaches beyond 12 months on reattempts has not yet been demonstrated. Also, because the number of trials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on death by suicide. in a randomized controlled trial of the post-crisis suicide prevention long-term follow?up contact approach, Motto and Bostrom (2001) found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow-up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and attempts {Hassanian-Moghaddam, Sarjami, Kolahi, 8: Carter, 2011; Wang et al., 2016i Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related difficulties) leading up to and following the suicidal event; safety plan development; skill building; and CST-SP also has family skill modules focused on family support and communication patterns as well as improving the family's problem solving skills. A randomized controlled trial of CST-SP found that 10-session outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide reattempt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual (Brown et al., 2005). 36 Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world. Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friend/peer, family member, co-worker, or other close contact to suicide (Pitman, Osborn, King, EL Erla ngsen, 2014}. Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide contagion {Etzersdorfer 3a Sonneck, 1998; Niederkrotenthaler E: Sonneck, 2007}. While the evidence is still being built in this area, particularly with regard to the impact of policy and practices on suicide and suicide attempts in the United States, measures to care forthe bereaved population through such means as postvention interventions counseling, support groups and debriefing sessions} and safe reporting on suicide have shown impacts in other countries. Approaches Some approaches that can be used to lesson harms and reduce future risk of suicide include caring for the bereaved and safe reporting following a suicide. - Postvention approaches are implemented after a suicide has taken place and may include debriefing sessions, counseling, and/or bereavement support groups for surviving friends and family membersfloved ones. These programs have not typically been evaluated for their impact on suicide or suicidal behavior but may reduce survivors? guilt, feelings of depression, and complicated grief (Szumilas Kutcher, 2011}. 0 Safe reporting following a suicide. The manner in which information on a recent suicide is communicated to the public leg. school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are betteinclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline}, and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion. Potential Outcomes I Reductions in ideation/attempts 37 I Reductions in distress I Increases in treatment seeking - Improvements in reporting following suicide - Reductions in contagion effects related to suicide Evidence Current evidence suggests that lessening harm through postvention and safe reporting can impact risk and protective factors for suicide. Postvention programs are implemented with the goal of providing support to survivors of others? suicide to reduce their own risk of suicide. One example of a postvention program, StondBy Response Service {5tondBy}, provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs {Vissen Comans, 8L Scuffham, 2014). In a study by Visser et al. (2014), StondBy clients were significantly less likely to be at high risk for suicidality {suicide ideation and attempts) and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program (48% and 64% respectively). Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention (versus passive approaches where survivors self-refer for services) {Cerel 8: Campbell, 2008). Safe reporting and messaging about suicide. One way to ensure safe reporting and messaging about suicide is to encourage news media adhere to Recommendations for Reporting on Suicide The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity of media reporting resulted in a nationwide significant reduction of 81 suicides annually (Niederkrotenthaler 8: Sonneck, 2007). Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide (Niederkrotenthaler et al., 2010). Reports of individual suicidal ideation not accompanied by reports ofsuicide or suicide attempts, along with reports describing a "mastery? of a crisis situation where adversities were overcome, was associated with significant decreases in suicide rates in the time period immediately following such reports {Niederkrotenthaler et al., 2010). 38 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy to Prevent Suicide, the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal}, social services, health services, business, labor, justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in Appendix A along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems), and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness of and garnering support for policies affecting individuals and families. The public health sector has been atthe forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teach Coping and Problem-Solving Skills to prevent suicide from happening in the first place. These programs are often delivered in school and community settings, making education and non-governmental organizations vital partners in prevention. 39 Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can play an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch oftheir state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. in a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identify and Support People ?it-Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation of this package. In this regard, all sectors can play an important and influential role in preventing suicide from happening in the first place and lessening the immediate and long-term harms of suicidal behavior by helping those in times of crisis get the services and support they need. 40 Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. it is important to have timely and reliable data to monitor the extent of the problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent data allow public health and other entities to better gauge the scope of the problem, identify high-risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying definitions of suicidal ideation, behavior, and death that can make it difficult to consistently monitor specific outcomes across sectors and over time. For example, the manner in which deaths are classified can change from one jurisdiction to another, and can change based on local medical andfor medicolegal standards (A.E. Crosby, Ortega, et al., 2011). uniform definitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems A.E. Crosby, Ortega, et al., 2011). Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. Notional Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of?death data from death certificates. is a state-based surveillance system (currently in 40 states, the District of Columbia, and Puerto Rico} that combines data from death certificates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches (Blair, Fowler, Jack, Crosby, 2016]. Data from state and local Child Death Review teams and Suicide Death Review Teams {which are in a few states] offer another source to identify deaths and obtain insight into the gaps in services, systems, and modifiable risk factors for suicide. The National Electronic lnj'ury Surveillance System-All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in U.S. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc], age, race/ethnicity, sex, disposition (where the injured person goes when released from the Emergency Department}. In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9-12 grade 41 students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose (Brener et al., 2013). The data are obtained from a national school-based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies. The l?llotiomtzlI Survey on Drug Use and Health (NSDUH) is an annual survey of the civilian, non- institutionalized population aged 12 years and older. NSDUH provides both national and state-level estimates of substance use (alcohol, tobacco, illicit drugs, and non?medical use of prescription drugs); mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groUps at increased risk. It is also important at all levels (local, state, and federal) to address gaps in responses, track progress of prevention efforts and evaluate the impact of those efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and long- term outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide {and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on riskfactors. More research is also needed to examine the effectiveness of primary prevention strategies (before risk occurs) and community?level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes. Conclusion Suicide is a serious public health problem whose rates have been on the rise for more than a decade and whose costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern. There are a number of barriers that have impeded progress, including, for example, stigma related to help-seeking, 42 mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is preventable, and more is being done to prevent suicide than ever before, as evidenced by the work of the National Action Alliance for Suicide Prevention, the release of the first world report on suicide, and more timely surveillance data, to name just a few examples. in an effort to continue pushing the field and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi- sectoral way. It includes strategies and approaches to prevent suicide from occurring in the first place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. Importantly, this technical package extends the bounds of the typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and financial security). While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. in keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be refined to reflect the current state of the science. in closing, and in keeping with a message of resilience as spoken by those with lived experience, 'hope, help, and healing is possible.? 43 Eieferences Archer, J., Bower, P., Gilhody, 5., Lovell, K., Richards, 0., Gask, L., . . . Coventry, P. (2012). Collaborative care for-"? depression and anxiety probIEms. The Cochrone Library. ArsenauIt-Lapierre, 6., Kim, C., St Turecki, G. (2004). diagnoses in 3215 suicides: a meta?analysis. BMC 4, 37. K. E., Canham?Chervak, M., Black, S. A., Dada. E. 0., Millikan, A. M., B. H. [2012). Mental health risk factors for suicides in the US Army, 2001?8. injury Prevention, 13(6), 405-412. Bandura, A. [1985). Social foundations of thought and action: A social cognitive theory: PrenticesHall, Inc. Barker, E., Kolves, K., St De Leo, D. [2014). Management of Suicidal and Self-Harming Behaviors in Prisons; Systematic Literature Review of Evidence-Based Activities. Archives ofSuicide Research, 13(3), Beautrals, A. L. {2001}. Effectiveness of barriers at suicide jumping sites: A case study. Aust N21 35. Beautrais, A. L., Gibb, 5.1., Fergusson, D. M., Horwood, L. J., St Larkin, G. L. {2009). Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZJ 43. Bellis, M. A., Hughes, K., Leckenby, N., Jones, L., Baban, A., Kachaeva, M., . . . Ulukoi, B. (2014). Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bulletin of the Wood Health Organization, 92(9). 641?655. Beyer, K. M., Layde, P. M., Hamberger, L. K., E: Laud, P. W. (2015). Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women, 21:1), 49-54. Blair, J. M., Fowler, K. A., Jack, 5. P., 8i. Crosby, A. E. [2015). The National Violent Death Reporting System: overview and future directions. inj Prev, 22 Suppll, i?-ill. Bossarte, R. M., Karras. E., Lu. N., Tu, it, Stephens, B., Draper, J., 3: Kemp, J. E. [2014). Associations between the Department of Veterans Affairsr suicide prevention campaign and calls to related crisis lines. Pubiic Health Rep, 129(6), 515?525. Branas, C. C., Cheney, R. A., MacDonald, J. M, Tam, W., Jackson, T. 0., &Ten Have, T. R. [2011). A difference? ln?dlfferences analysis of health, safety, and greening vacant urban space. Am Epidemiol, 1?4l11). 1236-1305. Brown, 6. K., Ten Have, T., Henriques, G. R., Hie, 5. it, Hollander, J. E., Ed. Beck, A. T. [2005). Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. JAMA, 294(5). 553-570. Bruce, M. L., Ten Have, T. R., Reynolds C. F., Katz, l. Schulberg, H. C., Mulsant, B. H., . . . Alexopoulos, G. S. [2004). Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA, 291119), 1031-1091. Capaldi, D. M., Knoble, N. E., Shortt, J. W., 8: Kim, H. K. [2012). ASystematic Review of Risk Factors for Intimate Partner Violence. Partner Abuse, 231?280. Carter, M., McGee R., Taylor, 5., Eu Williams, S. Health outcomes in adolescence: associations with family, friends and school engagement} Adolesc, 30(1), 51?62. Centers for Disease Control and Prevention. (2009]. Strategic direction for the prevention of suicidal behavior: Promoting individual, family. and community connectedness to prevent suicidal behavior. Retrieved from Available at: Strategic Direction Full versiondapdf Centers for Disease Control and Prevention. (2015). Web-Based Injury Statistics Query and Reporting System (WISQARS). Retrieved June 3, 2016 Available at: 44 Comment IA): Double?check EndNote- formatted references. There are several instances where the publication year and issuefpage numbers are incomplete. [Due to EndNote format, unable to specifically point out all of the issues in this document] There are also formatting issues in the text, especially with the insertion of the Lauthors initials in embedded citations. Cerel, J., Campbell, F. R. (2008). Suicide survivors seeking mental health services: a preliminary examination of the role of an active postvention model. Suicide Life Threat Behav, 38(1), 30-34. Cerel, 1., Maple, M., De Venne, A., Moore, M., Flaherty, C., Brown, M. (2016). Exposure to Suicide in the Community: Prevalence and Correlates in One US State. Pubiic Heaith Reports, 131(1). Cerel, J., McIntosh, i. L., Neimeyer, R. A., Maple, M., 8: Marshall, D. (2014). The Continuum of ?Survivorship?: Definitional Issues in the Aftermath of Suicide. Suicide and ?fe-threatening behavior, 44(6), 591-600. Chapman, A. L., 8: Dixon-Gordon, K. L. (2007). Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav, 37(5), 543-552. Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., 8: Anda, R. F. (2004]. Adverse childhood experiences and the risk of depressive disorders in adulthood. Affect Disord, 82(2), 217-225. Cherpitel, C. J., Borges, G. L. (3., E: Wilcox, H. C. (2004). Acute alcohol use and suicidal behavior: A review of the literature. Aicohoiisrn: Ciinicai and Experimentai Research, 28(5 SUPPL.), 135-285. Classen, T. J., 84 Dunn, R. A. (2012). The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Heaith Econ, 21(3), 338-350. Coffey, C. E. (2006). Pursuing perfect depression care. Serv, 52(10), 1524-1526. Coffey, C. E. (2007). Building a system of perfect depression care in behavioral health. it Comm Quai Patient Saf, 33(4), 193?199. Coffey, C. E., Coffey, M. J., 8i Ahmedani, B. K. (2013). An update on perfect depression care. Serv, 64(4), 396. Coffey, M., Coffey, C., 8: Ahmedani, B. K. {2015). Suicide in a health maintenance organization population. JAMA 72(3), 294?296. Comtois, K. A., Jobes, D. A., S. O'Connor, 5., Atkins, D. C., Janis, K., E. Chessen, C., . . . Yuodelis?Flores, C. (2011). Collaborative assessment and management of suicidality (CAMS): feasibility trial for next?day appointment services. Depress Anxiety, 28(11), 963-972. Cox, 6. R., Owens, C., Robinson, J., Nicholas, A., Lockley, A., Williamson, M., . . . Pirkis, i. (2013). Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Pubiic Heaith, 13(1), 1?12. Crosby, A. E., Espitia?Hardeman, V., Ortega, L., 8i. Loza no, B. (2013). Alcohol and suicide. Aicohoi: Science, Poiicy and Pubiic Heaith, 190. Crosby, A. E., Han, B., Ortega, L., Parks, S. E., Gfroerer, J. (2011). Suicidal Thoughts and Behaviors Among Adults Aged 218 Years United States, 2008?2009. MMWR, 60(No. 5513), 1-22. Crosby, A. E., Ortega, L., Melanson, C. (2011). Seif?directed Vioience Uniform Definitions and Recommended Data Eiements, Version 1.0. Retrieved from Atlanta, GA: Crosby, A. E., 8: Sacks, J. J. {2002). Exposure to Suicide: Incidence and Association with Suicidal Ideation and Behavior: United States, 1994. Suicide and iife-threatening behavior, 32(3), 321-328. Cunningham, P. i. (2009). Beyond parity: primary care physicians' perspectives on access to mental health care. Heaith Aff 28(3), w490-501. Curtin, S. C., Warner, M., St Hedegaard, H. (2016). increase in suicide in the United States, 1999-2014. Hyattsville, MD: National Center for Health Statistics Retrieved from Cylus, J., Glymour, M. M., Avendano, M. (2014). Do generous unemployment benefit programs reduce suicide rates? A state fixed-effect analysis covering 1963-2008. Am Epidemioi, 180(1), 45-52. doi:10.1093fajefkwu 106 45 Dahlberg, L. L., Krug, E. G. {2002). Violence-a global public health problem. In E. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, at. R. Lozano (Eds), World Report on Violence and Health (pp. 1-56). Geneva, Switzerland: World Health Organization. Deisenhammer, E. A., Ing, C. M., Strauss, R., Kemmler, G., Hinterhuber, H., 8: Weiss, E. M. (2009). The duration of the suicidal process: how much time is left for intervention between consideration and accomplishment of a suicide attempt? Clin 70(1), 19-24. Diamond, G. S., vlilintersteen, M. B., Brown, G. K., Diamond, G. M., Gallop, R., Shelef, K., 3-: Levy, 5. (2010). Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Child Adolesc 49(2), 122-131. Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., Giles, W. H. {2001). Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: findings from the Adverse Childhood Experiences Study. JAMA, 286(24), 3089-3096. Dunne, E. J., McIntosh, J. L., 8: Dunne-Maxim, K. (Eds). (1937}. Suicide andits aftermath: Understanding and counseling the survivors. New York: Norton. Durkheim, E. (1897/1951). Suicide: A study in sociology (JA Spaulding G. Simpson, trans). Glencoe, it: Free Press.(0riginal work published 1897). Edwards, R. W., Jumper-Thurma n, P., Plested, B. A., Getting, E. R., 8: Swanson, L. (2000). Community readiness: Research to practice. Journal of Community 28(3), 291-302. Escobedo, L. (5., 8L Ortiz, M. (2002). The relationship between liquor outlet density and injury and violence in New Mexico. Accla' Anal Prev, 34(5), 689-694. Etzersdorfer, E., Sonneck, G. (1998). Preventing suicide by influencing mass?media reporting: The Viennese experience, 1980?1996. Arch Suicide Res, 4. Felitti, J., Anda, R. F., Nordenberg, 0., Williamson, D. F., Spitz, A. M., Edwards, V., . . . Marks, J. S. (1998). Relationship of childhood abuse and household dysfunction to many ofthe leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. Am Prev Med, 14(4), 245?258. Ferdon, C., Crosby, A. E., Caine, E. D., Hindman J., Reed, G. A., 8: Iskander, J. (In press). CDC Public Health Grand Rounds: Preventing Suicide through a Comprehensive Public Health Approach. MMWR. Flavin, P., 8: Radcliff, B. (2009). Public policies and suicide rates in the American states. Social indicators Research, 90(2), 195-209. Fleischmann, A., Bertolote, J. M., Wasserman, D., De Leo, D., Bolhari, J., Botega, N. J., . . .Thanh, H. T. (2008). Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bulletin of the World Health Organization, 86(9), 2034709. doi:50042A 96862008000900014 [pii] Florence, C., Simon, T., Haegerich, T., Luo, F., St Zhou, C. (2015). Estimated lifetime medical and work-loss costs of fatal injuries-United States, 2013. MMWR: Morbidity and mortality weekly report, 64(38), 1074?1072. Fowler, K. A., Gladden, R. M., Vagi, K. J., Barnes, J., 81 Frazier, L. (2015). Increase in suicides associated with home eviction and foreclosure during the US housing crisis: findings from 16 National Violent Death Reporting System States, 2005-2010. Am Public Health, 105(2), 3114316. Frieden, T. R. (2014). Six components necessary for effective public health program implementation. Am Public Health, 104(1), 17-22. doi:10.2105lAJPH.2013.301608 Giesbrecht, N., Huguet, N., Ogden, L., Kaplan, M. S., McFarland, B. H., Caetano, R., . . . Nolte, K. B. (2015). Acute alcohol use among suicide decedents in 14 US states: impacts of off-premise and on-premise alcohol outlet density. Addiction, 110(2), 300-307. Gilbody, 5., Bower, P., Fletcher, J., Richards, D., 8: Sutton, A. J. (2006). Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives ofinternal medicine, 165(21), 2314-2321. 46 Godoy Garraza, L., Walrath, C., Goldston, D. 3., Reid, H., McKeon, R. (2015). Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on Suicide Attempts Among Youths. JAMA 72(11), 1143-1149. Gould, M. 5., Cross, W., Pisani, A. R., Munfakh, J. L., 8: Kleinman, M. {2013). Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav, 43(6), 676-691. Gould, M. S., Kalafat, J., Harrismunfakh, J. L., 8: Kleinman, M. {2007). An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav, 37(3), 338-352. Grossman, D. C., Mueller, B. A., Riedy, C., Dowd, M. D., Villaveces, A., Prodzinski, J., . . . Harruff, R. (2005). Gun storage practices and risk of youth suicide and unintentional firearm injuries. JAMA, 293(6), 3?07-714. Haddon, W. (1980). Advances in the epidemiology of injuries as a basis for public policy. Pubiic Heoith Reports, 95(5), 411-421. Haegerich, T. M., 8: Dahlberg, L. L. (2011). Violence as a public health risk. American Journai oftifestyie Medicine, 1559827611409127. Hailey, D., Roine, H., 81 Ohinmaa, A. (2008}. The effectiveness of telemental health applications: a review. The Canadian Journai of 53(11), 769-278. Hamby, 5., J. [2013). The web of violence: exploring connections among different forms of interpersonal violence and abuse Briefs in Socioiogy. New York, NY: Springer. Han, E., Crosby, A. E., Ortega, L. A., Parks, S. E., Compton, W. M., 84 Gfroerer, J. (2016). Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?E4years in the United States. Comprehensive 66, 176486. Harris, E. (2., 8L Barraclough, B. (1997). Suicide as an outcome for mental disorders. A meta~analysis. 120, 205-228. Harris, E. (2., 8L Barraclough, B. (1998). Excess mortality of mental disorder. Br.i 173, 11?53. Harris, K. M, Carpenter, (2., 8t Bao, Y. {2006). The effects of state parity laws on the use of mental health care. Med Care, 44(6), 499?505. Hassanian?Moghaddam, H., Sarjami, S., Kolahi, A. A., Carter, 6. L. (2011). Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital~treated selfnpoisoning. British Journai of 198(4), 309-316. Hawkins, J. D., Catalano, R. F., 8t. Kuklinski, M. R. (2014). Communities that ca re Encyciopeo?ia of Criminoiogy and Criminai Justice (pp. 393-408): Springer. Hawton, K. (2007). Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis, 28. Hayes, L. M. (1995). Prison Suicide: An Overview and a Guide to Prevention. The Prison Journai, 25(4), 431-456. Hayes, L. M. (2013). Suicide prevention in correctional facilities: reflections and next steps. int} Law 188?194. Herman, K. C., Borden, L. A., Reinke, W. M., 8: Webster-Stratton, C. (2011). The Impact ofthe Incredible Years Parent, Child, and Teacher Training Programs on Children's [Io-Occurring Internalizing 0, 26(3), 189-201. Houle, J. N., 8t Light, M. T. {2014). The home foreclosure crisis and rising suicide rates, 2005 to 2010. Am .i Puhiic Heaith, 104(6), 1073-1029. Hunkeler, E. M., Katon, W., Tang, L., Williams, J. W., Jr., Kroenke, K., Lin, E. H., . . . Unutzer, J. {2006). Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. 332(2536), 259-263. 47 lnagaki, M., Kawashima, Y., Kawanishi, C., Yonemoto, N., Sugimoto, T., Furuno, T., . . . Yamada, M. {2015). Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. .lAffect Disord, 175, 66-78. doi:10.1016fj.jad.2014.12.048 Jobes, D. A. (2012). The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence- based clinical approach to suicidal risk. Suicide Life Threat Behav, 42(6), 6140-653. doi:10.1111ij.1943- 278X.2012.00119.x Johnson, F. W., Gruenewald, P. J., 8: Remer, L. G. (2009). Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res, 33(12), 2124-2133. Kann, L., McManus, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Hawkins, J., . . .Zaza, S. {2016). Youth Risk Behavior Surveillance - United States, 2015. MMWR Surveill Summ, 65(6), 31?174. Kaplan, M. S., McFarland, B. H., Huguet, N., Conner, K., Caetano, FL, Giesbrecht, N., 3: Nolte, K. B. (2013). Acute alcohol intoxication and suicide: a gender-stratified analysis of the National Violent Death Reporting System. in} Prev, 19(1), 38-43. Kellam, S. (3., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., Toyinbo, P., . . . Wilcox, H. C. (2008). Effects of a universal classroom behavior management program in first and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend, 95 Suppl 1, 55-528. Kessler, R. (2., Demler, 0., Frank, R. 6., Olfson, M., Pincus, H. A., Walters, E. E., . . . Zaslavsky, A. M. (2005). Prevalence and treatment of mental disorders, 1990 to 2003. EnglJ Med, 352(24), 2515?2523. Kleiman, E. M., Riskind, J. H., Schaefer, K. E., 84 Weingarden, H. (2012). The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis, 33(5), 273-279. doi:10.102710227~ 5910/a000136 Klomek, A. E., Sourander, A., at Gould, M. (2010). The association of suicide and bullying in childhood to young adulthood: a review of cross?sectional and longitudinal research findings. The Canadian Journal of 55(5), 282-288. Knox, K. L., Litts, D. A., Talcott, G. W., Feig, J. C., 3: Caine, E. D. (2003). Risk ofsuicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. 327(7423), 1376. Knox, K. L., Pflanz, S., Talcott, G. W., Campise, R. L., Lavigne, J. E., Bajorska, A., . . . Caine, E. D. (2010). The US Air Force suicide prevention program: implications for public health policy. Am Public Health, 100(12), 2452-2453. doi:10.2105lajph.2009.159871 Lang, M. (2013). The impact of mental health insurance laws on state suicide rates. Health Econ, 22(1), 73-88. doi:10.1002ihec.1816 Leeb, R. T., Lewis, T., 82 Zolotor, A. J. (2011}. A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal ofth?estyle Medicine, 4454-468. Lineberry, T. W., 3; O'Connor, 5.5. Suicide in the US Army. Mayo Clinic Proceedings, 87(9), 821-373. Linehan, M. M., Comtois, K. A., Murray, A. M., Brown, M. 2., Gallop, R. J., Heard, H. L., . . . Lindenboim, N. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen 63(7), 75?-766. Losel, F., St Farrington, D. P. (2012). Direct protective and buffering protective factors in the development of youth violence. Am Prev Med, 43(2 Suppl 1), 58-523. Maimon, 0., Browning, C. R., 8: Brooks-Gunn, J. (2010). Collective efficacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav, 51(3), 307-324. 48 Mark, T. L., Shern, D. L., Bagalman, J. E., 8: Cao, Z. (2007). Ranking America's mental health: An analysis of depression across the states. Alexandria, VA: Mental Health America. McIntosh, W. L., Spies, E., Stone, D. M., Lokey, C. N., Trudeau, A. R., 8: Bartholow, B. (2016). Suicide Rates by Occupational Group - 17 States, 2012. MMWR Morb Mortal Wkly Rep, 65(25), 641-645. Miller, M., Warren, M., Hemenway, D., Azrael, D. (2015). Firearms and suicide in US cities. in} Prev, 21(e1), e116-119. Mishara, B. L. (1995). The impact of Suicide. New York: Springer. Mishara, B. L., 8; Martin, N. (2012). Effects of a comprehensive police suicide prevention program. Crisis: Journal of Crisis intervention 3: Suicide, 33(3), 162-168. Mitchell, A. M., Kim, Y., Prigerson, H. (3., 8L Mortimer-Stephens, M. (2004). Complicated grief in survivors of suicide. Crisis, 25(1), 12-18. Mohr, D. C., Vella, L., Hart, 5., Heckman, T., 8: Simon, G. (2008). The effect of telephone-administered on of depression and attrition: A meta-analysis. Clinical Science and Practice, 15(3), 243-253. Motto, J. A., 3: Bostrorn, A. G. (2001}. A randomized controlled trial of postcrisis suicide prevention. Serv, 52(6), 828-833. Muennig, P., Cohen, A. K., Palmer, A., Zhu, W. (2013). The relationship between five different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med, 85, 18?26. National Action Alliance for Suicide Prevention. (2014). Research Prioritization Task A prioritized research agenda for suicide prevention: An action plan to save lives. National Action Alliance for Suicide Prevention Workplace Task Force. {2015). Comprehensive Blueprint for Workplace Suicide Prevention. Washington, DC. Author. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. (2014). Suicide prevention and the clinical workforce: Guidelines for training. Washington, D.C.: Author. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. (2014). The Way Forward: Pathways to hope, recovery, and wellness with insights from lived experience. Washington, D.C. Author. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. (2015). Responding to Grief, Trauma, and Distress After a Suicide: U.S. National Guidelines. Washington, DC. Author. Niederkrotenthaler, T., 8r Sonneck, G. (2007). Assessing the impact of media guidelines for reporting on suicides in Austria: interrupted time series analysis. Aust Z.l 41. doi:10.1080l00043620701266680 Niederkrotenthaler, T., Voracek, M., Herberth, A., Till, B., Strauss, M., Etzersdorfer, E., . . . Son neck, G. (2010). Media and suicide. Papageno Werther effect. BMJ, 341, c5841. doi:10.1136lbmj.c5841 Olfson, M., Gerhard, T., Huang, C., 5., EL Stroup, T. S. (2015). Premature Mortality Among Adults With Schizophrenia in the United States. 72(12), 1172-1181. Owens, D. (2002). Fatal and non-fatal repetition of self-harm: Systematic review. British Journal of 181(3), 193?199. Perron, S., Burrows, S., Fournier, M., Perron, P. A., 8-1 Ouellet, F. (2013). Installation of a bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health, 103(7), 1235-1239. doi:10.2105iajph.2012.301089 Pirkis, J., Too, L. S., Spittal, M. 1., Krysinska, K., Robinson, 1., 84 Cheung, Y. T. D. (2015). Interventions to reduce suicides at suicide hotspots: a systematic review and meta-a nalysis. The Lancet 2(11), 994- 1001. Pitman, A., Osborn, D., King, M., Erlangsen, A. (2014). Effects of suicide bereavement on mental health and suicide risk. Lancet 1, 86-94. 49 Plested, B. A., Edwards, R. W., Jumper-Thurman, P. (2006). Community readiness: A handbook for successful change. Fort CO: Tri-Ethnic Center for Prevention Research. Pollock, L. R., 8: Williams, J. M. (2004). Problem-solving in suicide attempters. Med, 34(1), 163-167. Reid, M. J., Webster-Stratton, C., 8: Hammond, M. (2003). Follow-up of children who received the Incredible Yea rs intervention for oppositional-defiant disorder: Maintenance and prediction of 2-year outcome. Behavior Therapy, 34(4), 471-491. Rowhani-Rahba r, A., Simonetti, J. A., Rivara, F. P. (2016). Effectiveness of Interventions to Promote Safe Firearm Storage. Epidemioi Rev, 38(1), 111-124. Rubenstein, L. Chaney, E. F., Ober, 5., Felker, B., Sherman, 5. E., Lanto, A., 8; Viva?, 5. (2010]. Using evidence- based quality improvement methods for translating depression collaborative ca re research into practice. Systems, Heaith, 28(2), 91. Ru nyan, C. W., Becker, A., Brandspigel, 5., Barber, C., Trudeau, A., Novins, D. (2016). Lethal Means Counseling for Parents of Youth Seeking Emergency Care for Suicidality. West} Emerg Med, 17(1), 8-14. Russell, 5. T., 8L Joyner, K. (2001). Adolescent sexual orientation and suicide risk: Evidence from a national study. American Journai of Pubiic Heaith, 91(8), 1276-1281. Shepard, D. 5., Gurewich, D., Lwin, A. K., Reed, 6. A., Jr., Silverman, M. M. (2016). Suicide and Suicidal Attempts in the United States: Costs and Policy Implications. Suicide Life Threat Behav, 46(3), 352-362. Silverman, M. M., 81 Maris, R. W. (1995). The prevention of suicidal behaviors: An overview. Suicide and We? threatening behavior, 25(1), 10?21. Simon, 0. R., Swann, A. C., Powell, K. E., Potter, L. 3., Kresnow, M. J., 3: O'Carroll, P. W. (2001). Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav, 32(1 Suppl), 49?59. Spoth, R. L., Guyll, M, 81 Day, 5. K. (2002). Universal family~focused interventions in alcohol~use disorder prevention: cost-effectiveness and cost?benefit analyses of two interventions. Stud Aicohoi, 63(2), 219? 223. Stack, 5., 8t Wasserman, I. (2007). Economic strain and suicide risk: a qualitative analysis. Suicide and iife? threatening behavior, 37(1), 103-112. Stokes, M. L., McCoy, K. P., Abram, K. M., Byck, G. R., Teplin, L. A. (2015). Suicidal Ideation and Behavior in Youth in the Juvenile Justice System: A Review of the Literature. Correct Heaith Care, 21(3), 222?242. Sudak, H., Maxim, K., Carpenter, M. {2008). Suicide and stigma: a review ofthe literature and personal reflections. Acad 32(2), 136-142. doi210.1176/appi.ap.32.2.136 Szumilas, M., 8; Kutcher, S. (2011). Post-suicide intervention programs: a systematic review. Can .i Pubiic Heaith, 102(1), 13?29. Toosi, M. (2015). Laborforce projections to 2024: The iaborforce is growing, but siowiy. Washington, D.C.: Bureau of Labor Statistics. Turecki, G. (2014). Epigenetics and suicidal behavior research pathways. Am Prev Med, 47(3), 5144-5151. Tyrer, P., Reed, G. M., 8: Crawford, M. J. (2015). Classification, assessment, prevalence, and effect of personality disorder. Lancet, 385(9969), 717-725. U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016a). Designated Heaith Professionai Shortage Areas Statistics, As of November 4, 2016. Retrieved from U.S. Department of Health and Human Services Health Resources and Services Administrations. (2016b). NHSC Ciinician Retention: A Story of Dedication and Commitment. Retrieved from U.S. Office of the Surgeon General, 8: National Action Alliance for Suicide Prevention. (2012). 2012 Nationai Strategyfor Suicide Prevention: Goais and Objectives for Action. Washington, DC: HHS. 50 U.S. Public Health Service. (1999). The Surgeon General's Call to Action to Prevent Suicide. Washington, D.C. Unutzer, J., Tang, L., Oishi, S., Katon, W., Williams, J. W., Jr., Hunkeler, E., . . . Langston, C. (2006). Reducing suicidal ideation in depressed older primarv care patients. Geriatr Soc, 54(10), 1550-1556. Visser, V. S., Comans, T. A., 8: Scuffham, P. A. (2014). Evaluation Of The Effectiveness Of A Community-Based Crisis Intervention Program For People Bereaved By Suicide. Journal of Community 42(1), 19-28. doi:10.1002ljcop.21586 Walrath, C., Ga rraza, L. (3., Reid, H., Goldston, D. B., 8-: McKeon, R. (2015). impact of the Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health, 105(5), 986-993. Wang, Y. C., Hsieh, L. Y., Wang, M. Y., Chou, C. H., Huang(2016}. Coping Card Usage can Further Reduce Suicide Reattempt in Suicide Attempter Case Management Within 3-Month Intervention. Suicide Life Threat Behav, 46(1), 105-120. Wasserman, 0., Hoven, C. W., Wasserman, C., Wall, M., Eisenberg, R., Hadlaczkv, G., . . . Carli, V. {2014). School- based suicide prevention programmes: The SEYLE cluster-randomised, controlled trial. The Lancet, 1536-1544. Webster-Stratton, C., Hammond, M. (1997). Treating children with early-onset conduct problems: a comparison of child and pa rent training interventions. Consult Clin 55(1), 93-109. Webster-Stratton, C., Reid, M. J., 8: Hammond, M. (2001]. Preventing conduct problems, promoting social competence: a parent and teacher training partnership in head start. Clin Child 30(3), 283?302. Webster?Stratton, C. H., Reid, M. J., 81 Beauchaine, T. (2011). Combining parent and child training for young children with ADHD. Journal of Clinical Child Adolescent 40(2), 191?203. Webster?Stratton, C., Jamila Reid, M., 8: Stoolmiller, M. (2008). Preventing conduct problems and improving school readiness: evaluation of the incredible veers teacher and child training programs in high?risk schools. Journal of child and 49(5), 471?488. Wilcox, H. (2., Kellam, S. G., Brown, C. H., Poduska, J. M., lalongo, N. 5., Wang, W., St Anthony, J. C. (2008). The impact of two universal randomized first? and second~grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend, 95 Suppl 1, S6043. Wilkins, N., Tsao, B., Hertz, M., Davis, R., St Klevens, J. (2014). Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. World Health Organization. (2013). Global and Regional Estimates of Violence Against Women: Prevalence and Health Effects of intimate Partner Violence and NonaPartner Sexual Violence. Geneva: World Health Organization. World Health Organization. (2014). Suicide Prevention: A Global imperative. Geneva, Switzerland: WHO Press. vaan, P. A., Brown, C. H., LoMurrav, M., Schmeelkaone, K., Petrova, M., Yu, (1., . . . Wang, W. (2010). An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am Public Health, 100(9), 1653-1661. doi: Yip, P., Caine, E., Yousuf, 5., Chang, S.-S., Wu, K., 81 Chen, (2012). Means restriction for suicide prevention. Lancet, 379. 51 Appendix A: Summary of Strategies and Approaches to Prevent Suicide Strengthen economic supports Strengthen access and delivery of suicide care Create protective environments household financial Unemployment benefit programs Other income supports The Notionoi Neighborhood Stabilization of mental health conditions in health insurance Mentoi Health Parity Laws 1/ Reduce in undersenred areas Nationoi Heoith Service Corps Telementoi heoith (TMHJ Safer suicide care Henry Ford Perfect Depression Core {Pre- cursor to Zero Suicide Reduce access to lethal means at?rlsk intervening or suicide hot spots Safe storage practices gency nseiing on Access to Lethal Means ED CALM Organizational policies and culture [Comment Healthca re Government (local, state, Federal] Businessflabor Government (local, state, Federal] Government (local, state, Federal] Healthcare Social services Government (local, state]: Public Health [Comment Local only? I Best Available Evidence Business/Labor . Tog ether for Life ?f Justice Air Force 5UlCldE? Prevention Program Government (lo cal, state, Federal] Correctional suicide prevention 1" Community-based policies to reduce excessive alcohol use Government (local, state) 5 Alcohol outlet v" Bu sine 5 ab or Peer norm rams 93 Public Health Promote Sources ofStrength Education connectedness Communi en ement activities gag Public Health 5 Greening vacant urban spaces Government Social emotional learning programs Public Health Youth Aware of Mental Health Program v" v? Teach coping Education and problem- Good Behavior Game 4? solving skills . Parenting skill and tanninr relationship approaches Public Health The incredible Years Education 5 1 Strengthening Families 10-14 Best Available Evidence Identify and at-risk support people Gatekeeper training Public Health Applied Suicide intervention Skilis Training Healthcare Garret Lee Smith Federal Grant Program v? Crisis Intervention Public Health Soc'al Services National Suicide Prevention Lifeline v? I Treatment for people at risk of suicide improving Mood - Promoting Access to 1/ Collaborative Treatment Collaborative Assessment and Management Healthca re of Suicidality (CAMS) Dialectical Behavioral Therapy (DST) Somal Servlces Justice Attachment-Based Family Therapy v? Translating initiatives for Depression into Ejj?ective Solutions project Treatment to prevent re-attempts ED Brief intervention with Follow-up Visits Health ca re Active follow-up contact approaches Social services CBTfar Suicide Prevention 52 Best Available Evidence Postvention Lessen harms StondBv Response Service 1/ 1/ ea care and prevent future risk Safe reporting following a suicide Public Health 1/ Media Gurdelmes Media 1This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing specific activities. Preventing Suicide: A Technical Package of Policy, Programs, and Practices Preventing Suicide: A Technical Package of Policy, Programs, and Practices Developed by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, 2017 Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention Atlanta, Georgia Centers for Disease Control and Prevention Anne Schuchat, MD (RADM, USPHS), Acting Director National Center for injury' Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. (2017) Preventing SuiciderA Technical Package ofPoiicies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Preventing Suio'de: A Technical Package of Policy, Programs, and Practices Contents Acknowledgements 5 External Reviewers 5 Overview 7 Strengthen Economic SupportsiS Strengthen Access and Delivery of Suicide Care 19 Create Protective Environments 23 Promote Connectedness 27 Teach COping and Problem-Solving Skills 31 Identify and Support People at Risk 35 Lessen Harms and Prevent Future Risk41 Sector Involvement 43 Monitoring and Evaluation 45 Conclusion 47 References49 Appendix: Summary of Strategies and Approaches to Prevent Suicide 58 Preventing Suicide: I. Technical Package of Policy, Programs, and Practices 3? Acknowledgements We would like to thank the following individuals who contributed in speci?c ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but de?nitely not least, we extend our thanks and gratitude to all the external reviewers fortheir helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Of?ce for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindman Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Dan Reidenberg Suicide Awareness Voices for Education (SAVE) Christine Schuier National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention The experts above are listed with their af?liations at the time this document was reviewed. Preventing Suicide: mama Package of Pnlicy. Programs. and Prattires .- by Ge?y I Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at risk; and lessening harms and preventing future risk. The strategies represented in this package include those with a focus on preventing the risk of suicide in the ?rst place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the National Strategy forSuicide Prevention1 and the National Action Alliance for Suicide Prevention's priority to strengthen community?based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education,justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a speci?c risk factor or outcomeFTechnical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The ?rst component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the speci?c ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision?making in communities and states. Preventing Suicide is a Priority Suicide, as de?ned by the Centers for Disease Control and Prevention (CDC), is part ofa broader class of behavior called self-directed viol'ence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury? Self-directed violence may be suicidal or non-suicidai in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is de?ned as a non?fatal selfndirected and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs.? In 2015 (the most recent year of available death data}, suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes?" In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes ofdeath since 1975 in the U.S.7 Overall suicide rates increased 28% from 2000 to 2015.5 Suicide is a problem throughout the life span; it is the third leading cause of death for youth 10?14 years of age, the second leading cause of death among people 1544 and 25234 years of age; the fourth leading cause among people 35 to 44 years of age. the ?fth leading cause among people ages 45?54 and eighth leading cause among people 5434 years of age? Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the life span occurring among non-Hispanic American Indianz?Alaska Native and non-Hispanic White population groups. in 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.?3 Other population groups disproportionately impacted by suicide include middle?aged adults (whose rates increased 35% from 2000 to 2015, with Steep increases seen among both males and females aged 35?64 years?,? Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the ?rst time in decades)?; workers in certain occupational and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority peersu'? Suicides re?ect only a portion of the problem.'5 Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.15 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self- harm injuries, 27 who reported making a suicide attempt, and over 22?r who reported seriously considering suicideFr?? Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal in?uences that interact with one another, often over time."5 The social ecological model?encompassing multiple levels of focus from the individual, relationship, community, and societal?is a useful framework for viewing and understanding suicide risk and protective factors identi?ed in the literature.? Risk and protective factors for suicide exist at each level. For example, risk factors included:5 - Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants - Relationship level: high con?ict or violent relationships, sense of isolation and lack of social support, family! loved one's history of suicide, ?nancial and work stress - Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications] - Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide)? Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status.LS - Preventing Suicide: A Technical Package of Policy, Program, and Practices . Exposure to violence is associated increased risk of 1 depression, post-traumatic -eress disorder (PTSD), a icid and Protective factors, or those in?uences that buffer against the risk for suicide, can also be found across the different levels of the social ecological model. Protective factors identi?ed in the literature include: effective coping and problem-solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community, and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal means."5These protective factors can either counter a speci?c risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide attempts.2MB Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence?? Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide The effects of violence in childhood and adolescence can be observed decades later, including severe problems with ?nances, family, jobs, and stress factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove bene?cial in preventing suicide?? Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one's community,? school}2 family,33 caring adults,34r35and pro?social peers36 can enhance resilience and help reduce risk for suicide and other forms ofviolence. Preventing Suicide: A Technical Package of Policy, Program, and Practices The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals. families, and communities.?40 In an early study, Crosby and Sacks?H estimated that 7% of the U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide.They also estimated that for each suicide, 425 adults were exposed, or knew about the death.41 In a more recent study, in one state, Cerel et al?? found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide andfor having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long?term health and mental health consequences ranging from anger, guilt. and physical impairment, depending on the means and severity of the attempt.? Similarly, survivors of a loved one's suicide may experience ongoing pain and suffering including complicated grief,? stigma, depression, anxiety, post- traumatic stress disorder, and increased risk of suicidal ideation and suicide?il46 Less discussed but no less important, are the ?nancial and occupational effects on those left behind.47 The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work?loss costs alone.?7 Adjusting for potential under?reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self?directed violence to be approximately $93.5 billion in 201 3.43 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.43 The true economic costs are likely higher, as neither study included monetary ?gures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable.? While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or Rather, suicide prevention is best achieved by a focus across the individual. relationship, family, community, and societal-levels and across all sectors, private and public.?5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta?analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found signi?cant preventive effects on suicide; cl meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi- experimental design) evaluation study that found signi?cant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving bene?cial effects on multiple forms of violence; no evidence of harmful effects on speci?c outcomes or with particular subgroups; and feasibility of implementation in a U5. context ifthe program, policy, or practice has been evaluated in another country. Preventing ?technical ?drag! If Ml}. Programim Practise: 6 Within this technical package, some approaches do not yet have research evidence demonstrating impact on rates of suicide but instead are supported by evidence indicating impacts on risk or protective factors for suicide heip? seeking, stigma reduction, depression, connectedness). In terms ofthe strength ofthe evidence, programs that have demonstrated effects on suicidal behavior reductions in deaths, attempts) provide a higher-level of evidence, but the evidence base is not that strong in all areas. For instance, there has been less evaluation of community engagement and family programs on suicidal behavior. Thus, approaches in this package that have effects on risk or protective factors re?ect the developing nature of the evidence base and the use of the best available evidence at a given time. It is also important to note that there is often signi?cant heterogeneity among the programs, policies, or practices that fall within one approach or strategy in terms of the nature and quality of the available evidence. Not all programs, poli- cies. or practices that utilize the same approach are equally effective, and even those that are effective may not work across all populations.Taiioring programs and conducting more evaluations may be necessary to address different population groups. The evidence-based programs. practices, or policies included in the package are not intended to be a comprehensive list for each approach, but rather to serve as examples that have been shown to impact suicide or have bene?cial effects on risk or protective factors for suicide. Contextual and Cross-Cutting Themes One important feature ofthe package is the complementary and potentially synergistic impact ofthe strategies and approaches. The strategies and approaches included in this technical package represent different levels of the social ecology, with efforts intended to impact community and societal levels, as well individual and relationship levels. The strategies and approaches are intended to work in combination and reinforce each other to prevent suicide (see box on page 12). The strategies are arranged in order such that those strategies hypothesized to have the greatest potential for broad public health impact on suicide are included ?rst, followed by those that might impact subsets of the population persons who have already made a suicide attempt]. . like most public health . problems, suicide is preventable. - - Preventing Suicide: A Technical Package of Policy, Programs, and Practices 1 i it Strategy Preventing Suicide Approach Strengthen economic supports Strengthen household ?nancial security - Housing stabilization policies Strengthen access and delivery of suicide care Coverage of mental health conditions in health insurance policies Reduce provider shortages in underserved areas Safer suicide care through systems change Create protective environments - Reduce access to lethal means among persons at risk of suicide Organizational policies and culture Community-based policies to reduce excessive alcohol use Promote connectedness Peer norm programs Community engagement activities Teach coping and problem-solving skills Social-emotional learning programs Parenting skills and family relationship programs Identify and support people at risk Gatekeeper training Crisis intervention Treatment for people at risk of suicide Treatment to prevent re-attempts Lessen harms and prevent future risk Postvention Safe reporting and messaging about suicide It is important to note that these strategies are not mutually exclusive but each has an immediate focus. For instance, social?emotional learning programs, an approach under the Teach Coping and Problem-Solving Skills strategy, sometimes include components to change peer norms and the broader environment. The primary focus of these programs, however, is to provide children and youth with skills to resolve problems in relationships, school, and with peers, and to help youth address other negative in?uences substance use) associated with suicide. Preventing Suicide: ATechnical Package of Policy, Programs, and Practices The goal of this package is to stress the importance of comprehensive prevention efforts and to provide examples of effective programs addressing each level of the social ecology, with the knowledge that some programs, practices, and policies may impact multiple levels. Further, those that involve multiple sectors and that impact multiple levels of the social ecology are more likely to have a greater impact on the overall burden ofsuicide. Suicide ideation, thoughts, attempts. and deaths vary by gender, racefethnicity, age. occupation, and other important population characteristic591?:l Further, certain transition periods are also associated with higher rates ofsuicide transition from working into retirement, transition from active duty military status to civilian In fact, suicide risk can change along with dynamic risk factors. For example, individuals? coping skills may change during periods of crisis and heightened stress, limiting their normal ability to effectively solve problems and cope. Research indicates that suicide risk changes as a result of the number and intensity of key risk and protective factors experienced.52 Ideally, the availability of multiple strategies and approaches tailored to the social, economic, cultural, and environmental context of individuals and communities are desirable as they may increase the likelihood of removing barriers to supportive and effective care and provide opportunities to develop individual and community resilience.? identifying programs, practices, and policies with evidence of impact on suicide, suicide attempts, or bene?cial effects on risk or protective factors for suicide is only the ?rst step. In practice, the effectiveness of the programs, policies and practices identi?ed in this package will be strongly dependent on how well they are implemented, as well as the partners and communities in which they are implemented. Practitioners in the ?eld may be in the best position to assess the needs and of their communities and work with community members to make decisions about the combination of approaches included here that are best suited to their context. Data-driven strategic planning processes can help communities with this work.53'55 These planning processes engage and guide community stakeholders through a prevention planning process designed to address a community?s pro?le of risk and protective factors with evidence?based programs, practices, and policies. These processes can also be used to monitor implementation, track outcomes, and make adjustments as indicated by the data. The readiness of the program for broad dissemination and implementation availability of program materials, training and technical assistance) can also in?uence program effects. Implementation guidance to assist practitioners, organizations and communities will be developed separately. This package includes strategies where public health agencies are well positioned to bring leadership and resources to implementation efforts. it also includes strategies where public health can serve as an important collaborator strategies addressing community and societal level risks}, but where leadership and commitment from other sectors such as business, labor or health care is critical to implement a particular policy or program workplace policies; treatment to prevent re?attemptsi.The role of various sectors in the implementation of a strategy or approach in preventing suicide is described further in the section on Sectorinvolvement. In the sections that follow, the strategies and approaches with the best available evidence for preventing suicide are described. Wanting Suicide: Tedmlcal Package We. 1? Strengthen Economic Supports Rationale Studies from the US. examining historical trends indicate that suicide rates increase during economic recessions marked by high unemployment rates, job losses, and economic instability and decrease during economic expansions and periods marked by low unemployment rates, particularly for working-age individuals 25 to 64 years Economic and ?nancial strain, such asjob loss, long periods of unemployment, reduced income, dif?culty covering medical, food, and housing expenses, and even the anticipation of such ?nancial stress may increase an individual?s risk for suicide or may indirectly increase risk by exacerbating related physical and mental health problems.55 Buffering these risks can, therefore, potentially protect against suicide. For example, strengthening economic support systems can help people stay in their homes or obtain affordable housing while also paying for necessities such as food and medical care, job training, child care, among other expenses required for daily living. In providing this support, stress and anxiety and the potential for a crisis situation may be reduced, thereby preventing suicide. Although more research is needed to understand how economic factors interact with other factors to increase suicide risk, the available evidence suggests that strengthening economic supports may be one opportunity to buffer suicide risk. Approaches Economic supports for individuals and families can be strengthened by targeting household ?nancial security and ensuring stability in housing during periods of economic stress. Strengthening household ?nancial security can potentially buffer the risk ofsuicide by providing individuals with the ?nancial means to lessen the stress and hardship associated with a job loss or other unanticipated ?nancial problems.The provision of unemployment bene?ts and other forms of temporary assistance, livable wages, medical bene?ts, and retirement and disability insurance to help cover the cost of necessities or to offset costs in the event of disability, are examples of ways to strengthen household ?nancial security. Housing stabilization policies aim to keep people in their homes and provide housing options for those in need during times of?nancial insecurity. This may occur through programs that provide affordable housing such as through government subsidies or through other options available to potential homebuyers such as loan modi?cation programs, move-out planning, or ?nancial counseling services that help minimize the risk or impact of foreclosures and eviction. Potential Outcomes - Reductions in foreclosure rates . Reductions in eviction rates - Reductions in emotional distress - Reductions in rates of suicide Preventing Suicide: A Technical Package of Policy, Programs, and Fractious Evidence There is evidence suggesting that strengthening household ?nancial security and stabilizing housing can reduce suicide risk. Strengthen household ?nancial security. The Federal- Stote Unemployment insurance Program allows states to de?ne the maximum amount and duration of unemployment bene?ts that workers are entitled to receive after a job loss.59 An examination of variations in unemployment bene?t programs across states demonstrated that the impact of unemployment on rates of suicide was offset in those states that provided greater than average unemployment bene?ts {mean level: $7,990 per person in U.S. constant dollars}.The effects of unemployment bene?t programs were also consistent by sex and age group.59 Another U.S. study examining the link between unemployment and suicide rates using suicide data, length of unemployment (less than 5 weeks, 5-14 weeks, 15-26 weeks, and greater than 26 weeks), and job losses found that the duration of unemployment, as opposed to just the loss of a job, predicted suicide risk.m Together, these results suggest that not only should state unemployment bene?t programs be generous in their ?nancial allocations, but also in their duration. Other measures to strengthen household ?nancial security transfer payments related to retirement and disability insurance, unemployment insurance compensation, medical bene?ts, and other forms of family assistance) have also shown an impact on rates of suicide. A study by Flavin and Radcliffe? examined the impact of statesF per capita spending on transfer payments, medical bene?ts, and family assistance (Temporary Assistance to Needy Families and total state spending on suicide rates between 1990-2000, controlling for a number of suicide risk factors residential mobility, divorce rate, unemployment rate) at the state level. As per capita spending on total transfer payments, medical bene?ts, and family assistance increased there was an associated decrease in state suicide rates. In terms of lives saved, Flavin and Radcliff calculated the cost of reducing a state's suicide rate by a full point for the years studied.? At the national level, they estimated 3,000 fewer suicides would occur per year nationwide if every state increased its per capita spending on these types of Preventing'Suicirle: ATechni-i Package of Policy, Programs, and Practices Evidence suggest-5 that stre Marking . Smog/arm ?basing? housing can reduce suicide risk. assistance by $45 per year?" Although this was a correlational study, the results demonstrate the potential bene?ts of policies that reach particularly vulnerable individuals during periods of great need. More evaluation studies are needed to further understand the outcomes impacted by programs such as these. Housing stabilization policies. The Neighborhood was designed to help neighborhoods suffering from high rates of foreclosure and abandonment by slowing the deterioration of the neighborhoods and providing affordable housing options for low, moderate, and middle-income homebuyers. This program also offers ?nancial assistance to eligible individuals for the purchase of a new home. Although this program has not been rigorously evaluated for its impact on suicide outcomes, it addresses foreclosure and eviction, which are risk factors for suicide. A longitudinal analysis of annual data on suicides and foreclosures demonstrated that as the proportion of foreclosed properties increased in U.S. states, so did the state suicide rate, particularly among working-aged adults.?53 Another study of data from 16 US. states participating in the Notionoi Vioient Death Reporting System found that suicides precipitated by home foreclosures and evictions increased more than 100% from 2005 (before the housing crisis began] to 2010 [after it had peaked)? Most of these suicides occurred prior to the actual loss of the decedent's home. These ?ndings suggest that integrating suicide prevention resources, messaging, and referrals into ?nancial, foreclosure, and move-out planning and counseling services may help to prevent suicide. Preventing Suicide: it Technical Package of Policy, Programs, and Practices 3? m, 'm I, Strengthen Access and Delivery of Suicide (are Rationale While most people with mental health problems do not attempt or die by suicide?19 and the level of risk conferred by different types of mental illness previous research indicates that mental illness is an important risk factor for suicide?? State-level suicide rates have also been found to be correlated with general mental health measures such as Findings from the National Comorbidity Survey indicate that relatively few people in the US with mental health disorders receive treatment for those conditions.m Lack of access to mental health care is one of the contributing factors related to the underuse of mental health services?1 Identifying ways to improve access to timely, affordable, and quality mental health and suicide care for people in need is a critical component to prevention.S Additionally, research suggests that services provided are maximized when health and behavioral health care systems are set up to effectively and ef?ciently deliver such care."2 Apart from treatment bene?ts, these approaches can also normalize help?seeking behavior and increase the use of such services. Approaches There are a number of approaches that can be used to strengthen access and delivery of suicide care, including: Coverage of mental health conditions in health insurance policies. Federal and state laws include provisions for equal coverage of mental health services in health insurance plans that is on par with coverage for other health concerns mental health parity)? Bene?ts and services covered include such things as the number of visits, co- pays, deductibles, inpatienthutpatient services, prescription drugs, and hospitalizations. Ifa state has a stronger mental health parity law than the federal parity law, then insurance plans regulated by the state must follow the state parity law. lfa state has a weaker parity law than the federal parity law le.g., includes coverage for some mental health conditions but not others}, then the federal parity law will replace the state law. Equal coverage does not necessarily imply good coverage as health insurance plans vary in the extent to which bene?ts and services are offered to address various health conditions. Rather it helps to ensure that mental health services are covered on par with other health concerns. Reduce provider shortages in underserved areas. Access to effective and state-of?the-art mental health care is largely dependent upon the training and the size ofthe mental health care workforce. Over 85 million Americans live in areas with an insuf?cient number of mental health providers; this shortage is particularly severe among low- income urban and rural communities.M There are various ways to increase the number and distribution of practicing mental health providers in underserved areas including offering ?nancial incentives through existing state and federal programs loan repayment programs] and expanding the reach of health services through telephone, video and web?based technologies. Such approaches can increase the likelihood that those in need will be able to access affordable, quality care for mental health problems, which can reduce risk for suicide. Preventing Suicide: ll Technical Package of Policy. ngrarns. and Practice i, at a Safer suicide care through systems change. Access to health and behavioral health care services is critical for people at risk of suicide; however this isjust one piece of the puzzle. Care should also be delivered ef?ciently and effectively. More speci?cally, care should take place within a system that supports suicide prevention and patient safety through strong leadership, workforce training, systematic identi?cation and assessment of suicide risk, implementation of evidence-based treatments (see identify and Support People at Risk}, continuity of care, and continuous quality improvement. Care that is patient-centered and promotes equity for all patients is also of critical importance.? Potential Outcomes - Increased use of mental health services - Lower rates of treatment attrition . Reductions in depressive - Reductions in rates of suicide attempts - Reductions in rates of suicide Evidence There is evidence suggesting that coverage of mental health conditions in health insurance policies and improving access and the delivery of care can reduce risk factors associated with suicide and may directly impact suicide rates. Coverage of mental health conditions in health insurance policies. The National Survey on Drug Use and Health is a nationally representative survey of the US. population that provides data on substance use, mental health conditions, and service utilization?? Using data from this survey, Harris, Carpenter, and Bao? found that 12 months after states enacted mental health parity laws, self-reported use of mental healthcare services signi?cantly increased. Moreover, subsequent research by Lang"3 examined state mental health laws and suicide rates between 1990 and 2004 and found that mental health parity laws, speci?cally, were associated with an approximate 5% reduction in suicide rates.This reduction, in the 29 states with parity laws, equated to the prevention of 592 suicides per year.59 Reduce provider shortages in underserved areas. One example ofa program to improve access to mental health care providers is the National Health Service Corps which offers ?nancial incentives to attract mental/behavioral health clinicians to underserved areas.? Programs such as NHSC encourage individuals to work in the mental health profession in locations designated as Health Professional Shortage Areas in exchange for student loan debt repayment. A 2012 retention survey conducted by the Health Resources and Services Administration (HRSA), found that 61% of mental and behavioral health care providers continued to practice in designated mental health shortage areas after their four year commitment to the Although this program has not been evaluated for impact on suicide, it addresses access to care, which is a critical component to suicide prevention. Telemental Health (TMH) services refer to the use of telephone, video and web-based technologies for providing or care at a distance.? TMH can be used in a variety of settings outpatient clinics, hospitals, military treatment facilities] to treat a wide range of mental health conditions. It can also improve access to care for patients in isolated areas, as well as reduce travel time and expenses, reduce delays in receiving care, and improve satisfaction interacting with the mental health care system. A systematic review of TMH services found that services rated as high or good quality were effective in treating mental health conditions such as depression, schizophrenia, substance - 20 Preventing Suidrle: ?technical Package of Policy, Programs, and Practices Access to health and behavioral health care services is critical for people afsaicideabuse, and suicidal ideation and suicidef?g Further, Mohr and colleagues? conducted a meta?analysis examining the effect of delivered speci?cally via telephone and found that it signi?cantly reduced depressive in comparison to face-to-face also found that treatment attrition rates were signi?cantly lower among patients receiving telephone-administered compared to patients receiving face-to-face therapy?? Thus, TMH may not only offer improved access to mental health care, but it may also ensure continuity of care, and thereby further reduce the risk for suicide. Safer suicide care through systems change. Henry Ford Health System, which is a large health maintenance organization (HMO) in the state of Michigan, pioneered PerfectDepression Cares" the pro-cursor to what is now called Zero Suicide. The overall goal of Perfect Depression Care was to eliminate suicide among HMO members. More broadly, the goal of the program was to redesign delivery of depression care to achieve ?breakthrough improvement? in quality and safety by focusing on effectiveness, safety, patient centeredness, timeliness, ef?ciency, and equity among patients. The program screened and assessed each patient for suicide risk and implemented coordinated continuous follow-up care system wide.?51 An examination of the impact of the program found that there was a dramatic and statistically signi?cant decrease in the rate of suicide between the baseline years, 1999 and 2000, and the intervention years, 2002-2009. During this time period, the suicide rate fell by Further, among HMO members who received mental health specialty services, the suicide rate signi?cantly decreased over time from 1999 to 2010 (110.3 to 47.6 per 100,000 population; p<.04) with a mean of 36.2 per 100,000 over the period.33 Additionally, for those HMO members who accessed only general medical services as opposed to specialty mental health services, the suicide rate increased from 2.7 to 5.6 per 100,000 Similarly, in the state of Michigan, rates of suicide in the general population increased over the period from 9.8 to 12.5 per Preventing Suicide: A Technical Package of Policy, Frog ram, and Practices 3? Create Protective Environments Rationale Prevention efforts that focus not only on individual behavior change help-seeking, treatment interventions) but on changes to the environment can increase the likelihood of positive behavioral and health outcomes.Em Creating environments that address risk and protective factors where individuals live, work, and play can help prevent suicide."17 For example, rates of suicide are high among middle-aged adults who comprise 42.6% ofthe workforce55; among certain occupational groupsm'?; and among people in detention facilities jail, prisonlfl?a to name a few. Thus, settings where these populations work and reside are ideal for implementing programs, practices and policies to buffer against suicide. Changes to organizational culture through the implementation of supportive policies, for instance, can change social norms, encourage help-seeking, and demonstrate that good health and mental health are valued and that stigma and other risk factors for suicide are notam Similarly, modifying the characteristics ofthe physical environment to prevent harmful behavior such as access to lethal means can reduce suicide rates, particularly in times of crisis or transition.39?94 Approaches The current evidence suggests three potential approaches for creating environments that protect against suicide. Reduce access to lethal means among persons at risk of suicide. Means of suicide such as ?rearms, hangingr? suffocation, orjumping from heights provide little opportunity for rescue and, as such, have high case fatality rates about 85% of people who use a ?rearm in a suicide attempt die from their Research also indicates that: 1) the interval between deciding to act and attempting suicide can be as short as 5 or 10 and 2) people tend not to substitute a different method when a highly lethal method is unavailable or dif?cult to accessf?ir?i9 Therefore, increasing the time interval between deciding to act and the suicide attempt, for example, by making it more dif?cult to access lethal means, can be lifesaving. The following are examples of approaches reducing access to lethal means for persons at risk of suicide: - intervening otSuicr'de Hotspots. Suicide hotspots, or places where suicides may take place relatively easily, include tall structures bridges, cliffs, balconies, and rooftops), railway tracks, and isolated locations such as parks. Efforts to prevent suicide at these locations include erecting barriers or limiting access to prevent jumping, and installing signs and telephones to encourage individuals who are considering suicide, to seek help.? - Safe Storage Practices. Safe storage of medications, ?rearms, and other household products can reduce the risk for suicide by separating vulnerable individuals from easy access to lethal means. Such practices may include education and counseling around storing ?rearms locked in a secure place leg, in a gun safe or lock box), unloaded and separate from the ammunition; and keeping medicines in a locked cabinet or other secure location away from people who may be at risk or who have made prior Organizational policies and culture that promote protective environments may be implemented in places of employment, detention facilities, and other secured environments residential settings). Such policies and cultural values encourage leadership from the top down and may promote prosocial behavior asking for help), skill building, positive social norms, assessment, referral and access to helping services mental health, substance abuse treatment, ?nancial counseling), and development of crisis response plans, postvention and other measures to foster a safe physical environment. Such policies and cultural shifts can positively impact organizational climate and morale and help prevent suicide and its related risk factors depression, social isolationiFfrmi Community-based policies to reduce excessive alcohol use. Research studies in the United States have found that greater alcohol availability is positively associated with alcohol?involved suicidesm'm Policies to reduce excessive alcohol use broadly include zoning to limit the location and density of alcohol outlets, taxes on alcohol, and bans on the sale of alcohol for individuals under the legal drinking age.?35 These policies are important because acute alcohol use has been found to be associated with more than one-third of suicides and approximately 40% of suicide attempts.?5 Potential Outcomes - Increases in safe storage of lethal means - Reductions in rates of suicide - Reductions in suicide attempts - Increases in help-seeking . Reductions in alcohol-related suicide deaths Evidence The evidence suggests that creating protective environments can reduce suicide and suicide attempts and increase protective behaviors. Reduce access to lethal means among persons at risk of suicide. A meta-analysis examining the impact of suicide hotspot interventions implemented in combination or in isolation, both in the U5. and abroad, found associated reduced rates of For example, after erecting a barrier on the Jacques?Cartier bridge in Canada, the suicide rate from jumping from the bridge decreased from about 10 suicide deaths per year to about 3 deaths per year.?is Moreover, the reduction in suicides byjumping was sustained even when all bridges and nearbyjumping sites were considered, suggesting little to no displacement of suicides to otherjumping sites.mg Further evidence for the effectiveness of bridge barriers was demonstrated by a study examining the impact of the removal of safety barriers from the Grafton Bridge in Auckland, New Zealand. After removal of the barrier, both the number and rate of suicide increased ?vefold-??11?? Another form of means reduction involves implementation of safe storage practices. In a case-control study of ?rearm-related events identi?ed from 37 counties in Washington, Oregon, and Missouri, and from 5 trauma centers, researchers found that storing ?rearms unloaded, separate from ammunition, in a locked place or secured with a safety device was protective of suicide attempts among adolescents!? Further, a recent systematic review of clinic and community-based education and counseling interventions suggested that the provision of safety devices signi?cantly increased safe ?rearm storage practices compared to counseling alone or compared to the provision of economic incentives to acquire safety devices on one's own.101 Preventing Suicide: A Technical Package of Policy, Program, and Practices 0 Another program, the Emergency Department Counseling on Access to Lethal Means (ED trained emergency clinicians in a large children?s hospital to provide lethal means counseling and safe storage boxes to parents of patients under age 13 receiving care for suicidal behavior. In a pre-post quality improvement project, Runyan et al39 found that at post-test 76% (of the 55% of parents followed up, n:114} reported that all medications in the home were locked up as compared to fewer than 10% at the time of the initial emergency department visit. Among parents who indicated the presence of guns in the home at pre-test all (100%) reported guns were currently locked up at post-test.Bg Organiaational policies and culture. Together for Life is a workplace program of the Montreal Police Force implemented to address suicide among of?cers. Policy and program components were designed to foster an organizational culture that promoted mutual support and solidarity among all members of the Force. The program included training of supervisors, managers and all units to improve competencies in identifying suicidal risk and to improve use and awareness of existing resources. The program also included an education campaign to improve awareness and help-seeking.? Police suicides were tracked over 12 years and compared to rates in the control city of Quebec.The suicide rate in the intervention group decreased signi?cantly by to a rate of 6.4 suicides per 100,000 population per year compared to an 11% increase in the control city {29.0 per Another example of this approach is the United States Air Force Suicide Prevention Program. The program included 1 1 policy and education initiatives and was designed to change the culture of the Air Force surrounding suicide.The program uses leaders as role models and agents of change, establishes expectations for behavior related to awareness of suicide risk, develops population skills and knowledge education and training], and investigates every suicide outcomes measurement]. The program represents a fundamental shift from viewing suicide and mental illness solely as medical problems and instead sees them as larger service?wide problems impacting the whole community.? Using a time?series design to examine the impact of the program on various violence?related outcomes, researchers found that the program was associated with a 33% relative risk reduction in suicide.? The program was also associated with relative risk reductions in related outcomes including moderate and severe family violence (30% and 54%. respectively}, homicide and accidental death A longitudinal assessment of the program over the period 1981 to 2008 (16 years before the 1997 launch of the program and 11 years post-launch] found signi?cantly lower rates ofsuicicle after the program was launched than before.? These effects were sustained over time, except in 2004, which the authors found was associated with less rigorous implementation of program components in that year than in the other years.? Finally, while the evidence is still being built for suicide prevention in correctional facilities, preliminary evidence suggests organizational policies and practices that include routine suicide prevention training for all staff; standardized intake screening and risk assessment; provision of shared information between staff members (especially in transitioning or transferring of inmates); varying levels of observation; safe physical environment; emergency response protocols; noti?cation of suicidal behavior/suicide through the chain of command; and critical incident stress debrie?ng and death review can potentially reduce suicide)? When these policies and practices were implemented across 1 1 state prisons in Louisiana, suicide rates dropped 46%, from a rate of 23.1 per 100,000 before the intervention to 12.4 per 100,000 the following year.?3 Similar programs have seen declines in suicide both in the United States and in other countries.? Community-based policies to reduce excessive alcohol use. While multiple policies to limit excessive use ofalcohol exist, several studies on alcohol outlet density and risk factors for suicide, such as interpersonal violence and social connectedness,?5??1 suggest that measures to reduce alcohol outlet density can potentially reduce alcohol-involved suicides. Additionally, a longitudinal analysis of alcohol outlet density, suicide mortality, and hospitalizations for suicide attempts over 6 years in 581 California zip codes, indicated that greater density of bars, speci?cally, was related to greater suicide and suicide attempts, particularly in rural areas.119 PmmtingSuldde: Package . ?it '1 3 Promote Connectedness Rationale Sociologist, Emile Durkheim theorized in 189? that weak social bonds, lack of connectedness, were among the chief causes of suicidalitme Connectedness is the degree to which an individual or group of individuals are socially close, interrelated, or share resources with others.?21 Social connections can be formed within and between multiple levels of the social ecology,? for instance between individuals peers, neighbors, co-workers}, families, schools, neighborhoods, workplaces, faith communities, cultural groups, and society as a whole. Related to connectedness, social capital refers to a sense of trust in one?s community and neighborhood, social integration, and also the availability and participation in social organizations?;123 Many ecological cross-sectional and longitudinal studies have examined the impact of aspects of social capital on depression depressive disorder, mental health more generally, and suicide. While the evidence is limited, existing studies suggest a positive association between social capital (as measured by social trust and community/neighborhood engagement], and improved mental Connectedness and social capital together may protect against suicidal behaviors by decreasing isolation, encouraging adaptive coping behaviors, and by increasing belongingness, personal value, and worth, to help build resilience in the face of adversity. Connectedness can also provide individuals with better access to formal supports and resources, mobilize communities to meet the needs ofits members and provide collective primary prevention activities to the community as a whole.121 Approaches Promoting connectedness among individuals and within communities through modeling peer norms and enhancing community engagement may protect against suicide. Peer narm programs seek to normalize protective factors for suicide such as help-seeking, reaching out and talking to trusted adults, and promote peer connectedness. By leveraging the leadership qualities and social in?uence of peers, these approaches can be used to shift group-level beliefs and promote positive social and behavioral change. These approaches typically target youth and are delivered in school settings but can also be implemented in community settings)? Community engagement activities Community engagement is an aspect of social capital? Community engagement approaches may involve residents participating in a range of activities, including religious activities, community clean-up and greening activities, and group physical exerciseThese activities provide opportunities for residents to become more involved in the community and to connect with other community members, organizations, and resources, resulting in enhanced overall physical health, reduced stress, and decreased depressive thereby reducing risk of suicide. . Preventing Suicide: A Technical Package of Policy. Program. am! Prank: 3F 0 Potential Outcomes Increases in healthy coping attitudes and behaviors Increases in referrals for youth in distress Increases in help?seeking behaviors Increases in positive perceptions of adult support Promoting connectedness among individuals and within -. communities may protect against suicide. 4?3- .- . - 28 Preventing Suicide: ATechnical Package of Policy, Programs, and Practices -r - .- Evidence Current evidence suggests a number of positive bene?ts of peer norm and community engagement activities, although more evaluation research is needed to examine whether these improvements in factors that protect against suicidal behavior translate into reduced suicide attempts and deaths. Peer norm programs. Evaluations show that programs such as Sources ofStrength can improve school norms and beliefs about suicide that are created and disseminated by student peers. In a randomized controlled trial of Sources of Strength conducted with 18 high-schools [6 metropolitan, 12 rural), researchers found that the program improved adaptive norms regarding suicide, connectedness to adults. and school engagement.36 Peer leaders were also more likely than controls to refer a suicidal friend to an adult. For students, the program resulted in increased perceptions of adult support for suicidal youths. particularly among those with a history of suicidal ideation, and the acceptability of help?seeking behaviors. Finally, trained peer leaders also reported a greater decrease in maladaptive coping attitudes compared with untrained leaders.35 Community engagement activities. A vacant lot greening initiative was undertaken in Philadelphia between 1999 and 2008. Local residents and community members worked together to green 4,436 lots (or 18 million square feet) in 4 areas of the city. Researchers found signi?cant reductions in community residents*self?reported level of stress, a risk factor for suicide. and engagement in more physical exercise. a protective factor for suicide, than residents in control vacant lot areas. There is some evidence for other cross-cutting bene?ts, including reductions in ?rearm assaults and vandalismPE-m Preventing Suicide: A Technical Package of Policy, Programs, and Practices 3? Teach Coping and Problem-Solving Skills Rationale Building life skills prepares individuals to successfully tackle every day challenges and adapt to stress and adversity. Life skills encompasses many concepts, but most often include coping and problem-solving skills, emotional regulation, con?ict resolution, and critical thinking. Life skills are important in protecting individuals from suicidal behaviorsu?i? Suicide prevention programs that focus on life and social skills training are drawn from social cognitive theories,13g surmising that suicidal behavior is attributed to either direct learning and modeling or environmental and individual hopelessness) characteristics. The inability to employ adequate strategies to cope with immediate stressors or identify and ?nd solutions for problems has been characterized among suicide attempters?l Teaching and providing youth with the skills to tackle every day challenges and stressors is, therefore, an important developmental component to suicide prevention. Approaches Social-emotional learning programs and parenting skill and family relationship programs are two approaches for teaching coping and problem-solving skills. Sociahemotional learning programs focus on developing and strengthening communication and problem?solving skills, emotion regulation, con?ict resolution, help seeking and coping skills. These approaches address a range of risk and protective factors for suicidal behavior. They provide children and youth with skills to resolve problems in relationships, school, and with peers, and help youth address other negative in?uences substance use) associated with suicide.I26 These approaches are typically delivered to all students in a particular grade or school, although some programs also focus on groups of students considered to be at high risk for suicide. Opportunities to practice and reinforce skills are an important part of programs that work.?32 Parenting skill and family relationship programs provide caregivers with support and are designed to strengthen parenting skills, enhance positive parent-child interactions, and improve children's behavioral and emotional skills and abilities.132 Programs are typically designed for parents or caregivers with children in a speci?c age range and can be self?directed or delivered to individual families or groups offamilies. Some programs have sessions primarily with parents or caregivers while others include sessions for parents or caregivers, youth, and the family. Speci?c program content typically varies by the age of the child but often has consistent themes of child development, parent?child communication and relationships, and youth?s interpersonal and problem-solving skills. Preventing Suicide: I. Technical Package of Policy, Programs, and Fraction 39F 3 Potential Outcomes - Reductions in suicide ideation - Reductions in suicide attempts - Reductions in suicide risk behaviors depression. anxiety, conduct problems, substance abuse] - Improvements in help-seeking behavior - Improvements in social competence and emotional regulation skills - Improvements in problem-solving and con?ict management skills Evidence Several social?emotional learning and parenting and family relationship programs have been shown in rigorous evaluations to improve resilience and reduce problem behavior and risk factors for various behaviors, including ones closely related to suicide, such as depression, internalizing behaviors, and substance abuse.133 Social-emotional learning programs. The Youth Aware ofMental Health Program (YAM) is a program developed for teenagers aged 14?16 that uses interactive dialogue and role-playing to teach adolescents about the risk and protective factors associated with suicide (including knowledge about depression and anxiety) and enhances their problem?solving skills for dealing with adverse life events, stress, school and other problems.134 In a cluster- randomized controlled trial conducted across 10 European Union countries and 168 schools, students in schools randomized to YAM were signi?cantly less likely to attempt suicide and have severe suicidal ideation at the 12+month follow?up compared to students in control schools which received educational materials and care as usual. Overall, the relative risk of youth suicide attempts among the YAM group was reduced by over 50% demonstrating that out of 1000 students, ?ve attempted suicide in the YAM group compared to in the control group. Additionally, related to severe suicide ideation, in the YAM group, relative risk fell by Another example is the Good Behavior Game which is a classroom-based program for elementary school children aged 6?10.The program uses a team-based behavior management strategy that promotes good behavior by setting clear expectations for good behavior and consequences for maladaptive behavior. The goal of the 636 program is to create an integrated classroom social system that is supportive of all children being able to learn with little aggressive or disruptive behavior.?35Two cohorts of youths participated in the program in 1985-86 and 198687 school years when they were in the ?rst and second grades. A number of proximal and distal outcomes were assessed among the two cohorts over time. With respect to distal suicide?related outcomes, an outcome evaluation of the 686 indicated that individuals in the ?rst cohort, who were assigned to participate in 636 when they were in the ?rst grade, reported half the adjusted odds of suicidal ideation and suicide attempts when assessed approximately 15 years later, between the ages of 19 to 21, compared to peers who had been in a standard classroom setting.The bene?cial effect ofthe program was consistent for suicidal ideation regardless ofwhether baseline covariates were included.?35The GBG effect on attempts was less robust in some adjusted models including caregiver mental health. In the second cohort of 686 students, neither suicidal ideation nor suicide attempts were signi?cantly different between 686 and the control interventions.?35The researchers believed this may have been due to a lack of implementation ?delity, including less mentoring and monitoring of teachers. 636 was also found to be associated with reduced risk of later substance abuse and other suicide risk factors among the ?rst cohort of students. Results for the second cohort were generally smaller but in the desired - 31 PreventingSuidde: ATetl'm'lal hckageorro?o. Programmeraaim 6 Parenting skill and family relationship programs. Parenting and family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. For example, the incredible Years (W) is a comprehensive group training program for parents, teachers and children designed to reduce conduct and substance abuse problems {two important suicide risk factors in youth} by improving protective factors such as responsive and positive parent-teacher?child interactions and relationships, emotion self-regulation and social competence {all protective factors for suicide).132 The program includes 9-20 sessions offered in community-based settings religious, recreation centers, mental health treatment centers, and hospitals). Several studies have demonstrated the effect of the program on reducing internalizing such as anxiety and depression, and child conduct problems.131133 The program is also associated with improved problem-solving and conflict management; these skills were maintained at 1~year Additionally, the program demonstrated greater bene?ts in mother?rated child internalizing compared to the waitlisted control group, when parent, child, and teacher components were included.132 Additionally, Strengthening Families i0? i4 is a program that involves sessions for parents, youth, and families with the goal of improving parents? skills for disciplining, managing emotions and con?ict, and communicating with their children; promoting youths?interpersonal and problem-solving skills; and creating family activities to build cohesion and positive parent-child interactions. The premise of the program is that developing these skills for both parents and children will reduce internalizing behavior and adolescent substance abuse, two important risk factors for suicide.142 Strengthening Families has been shown to signi?cantly decrease externalizing behaviors, such as aggression, alcohol use, and drug use among youth participants, as well as reduce depression, alcohol use, and drug use among participating families.?42 Parenting and a family skills training approaches have shown promising impacts in preventing key risk factors associated with suicide. Y- Preventing Suicide: ll Technical Package of Policy, Programs, and Practices 3? . ll Identify and Support People at Risk Rationale In order to decrease suicide, care of, and attention to, vulnerable populations is necessary, as these groups tend to experience suicidal behavior at higher than average rates. Such vulnerable populations include, but are not limited to, individuals with lower socio-economic status or who are living with a mental health problem; people who have previously attempted suicide; Veterans and active duty military personnel; individuals who are institutionalized, have been victims of violence, or are homeless; individuals of sexual minority status; and members of certain racial and ethnic minority Supporting people at risk requires proactive case ?nding and effective response. crisis intervention, and evidence-based treatment. Finding optimal ways of identifying at risk individuals, customizing services to make them more accessible Internet?based services when appropriate] and engaging people in evidence? based care through such measures as collaborative treatment}, remain key challengesfi-?t?r?t5 Simply improving or expanding services does not guarantee that those services will be used by people most in need, nor will it necessarily increase the number of people who follow recommended referrals or treatment. For example, some people living in disadvantaged communities may face social and economic issues that can adversely affect their ability to access supportive services.m App roaches The following approaches focus on identifying and supporting people at increased risk of suicide. Gatekeeper training is designed to train teachers, coaches, clergy, emergency responders, primary and urgent care providers, and others in the community to identify people who may be at risk of suicide and to respond effectively, including facilitating treatment seeking and support services. Gatekeeper training may be implemented in a variety of settings to identify and support people at risk.?15 Crisis intervention These approaches provide support and referral services, typically by connecting a person in crisis {or a friend or family member of someone at risk) to trained volunteers or professional staff via telephone hotline, online chat, text messaging, or in-person. Crisis intervention approaches are intended to impact key risk factors for suicide, including feelings of depression, hopelessness, and subsequent mental health care utilization.m Similar to means reduction, crisis interventions can put space or time between an individual who may be considering suicide and harmful behavior. Treatment for people at risk of suicide can include various forms of delivered by licensed providers to help individuals with mental health problems and other suicide risk factors with and emotional regulation. Treatment usually takes place in a one-on?one or group format between patients and clinicians and can vary in duration from several weeks to ongoing therapy, as needed. Treatment that employs collaborative lie, between patient and therapist or care manager) and/or integrated care linkage between primary care and behavioral health care) can help engage and motivate patients, thereby increasing retention in therapy and decreasing suicide risk-148450 Preventing Suicide: A Technical Package of Policy. Program. and Pram-u Treatment to prevent reattempts. These approaches typically include follow?up contact and use diverse modalities ie.g., home visits, mail, telephone, e?mail) to engage recent suicide attempt survivors in continued treatment to prevent Treatment may focus on improved coping skills, mindfulness, and other emotional regulation skills, and may include case management home visits to increase adherence to treatment and continuity of care; and one-on-one interpersonal therapy andior group therapy. Approaches that engage and connect people who have attempted to peers and providers are especially important because many attem pters do not present to aftercare; 12%?25% re?attempt within a year, and 396?996 of attempt survivors die by suicide within I to 5 years of their initial attemptm Potential Outcomes - Reductions in suicidal ideation - Reductions in suicide attempts - Reductions in suicide rates - Reductions in depression and feelings of hopelessness - Reductions in re?attempts - Improvements in coping skills . Increases in treatment engagement and compliance with medications Evidence The current evidence suggests that identifying people at risk of suicide and the continued provision oftreatment and support for these individuals can positively impact suicide and its associated risk factors. {jatekeeraer training Applied Suicide intervention Training is a widely implemented training program that helps hotline counselors, emergency workers, and other gatekeepers to identify and connect with suicidal individuals, understand their reasoning for living and dying, and assist with safely connecting those in need to available resources. In a study employing a randomized controlled trial, Gould, Cross, Pisani, Munfakh, 8: Kleinman?? evaluated the training across the NotionaiSuicide Prevention tifeiine network of hotlines over the period 2008-2009. Using data from 1,41 0 suicidal individuals who called 17 Lifeline centers, the researchers found that callers who spoke with ASiST?trained - 7m WNW 0 counselors were signi?cantly more likely to feel less depressed, less suicidal, less overwhelmed, and more hopeful by the end oftheir call, compared to callers who spoke to non~ASiSTtrained counselors. Counselors trained in ASIST were also more skilled at keeping callers on the phone longer and establishing a connection with them. However, training in ASiSTdid not result in more comprehensive suicide risk assessments than usual care training?: Gatekeeper training has also been a primary component of the Garret Lee Smith (GL5) Suicide Prevention Program, which has been implemented in 50 states and 50 tribes. A multi-site evaluation assessed the impact of community gatekeeper training on suicide attempts and deaths by comparing the change in suicide rates and nonfatal suicidal behavior among young people aged 10?24 in counties implementing GL5 trainings, with the trajectory observed in similar counties that did not implement these trainings. Counties that implemented GLS trainings had signi?cantly lower youth suicide rates one year following the training implementationJ-?This ?nding equates to a decrease of 1 suicide death per 100,000 youth ages 10 to 24, or the prevention of approximately 237 deaths in the age group, between 2007 and 2010. Counties implementing GL5 program activities also had signi?cantly lower suicide attempt rates among youth ages 16 to 23 in the year following implementation of the GL5 program than did similar counties that did not implement GL5 activities (4.9 fewer attempts per 1000 More than 29,000 suicide attempts may have been prevented during the period examined. Crisis intervention. Suicide prevention hotlines are one way to provide crisis intervention. In an evaluation of the effectiveness of the Notionni Suicide Prevention Lifeiine to prevent suicide, 1,085 suicidal individuals who called the hotline completed a standard risk assessment for suicide, and 380 of those completed a follow-up assessment between 1 and 52 days (mean=13.5 days) after the initial assessment. Researchers found that over half ofthe initial sample were seriously considering suicide when they called, and they had a plan for their suicide. Researchers also found that among follow-up participants, there was a signi?cant decrease in pain, hopelessness, and intent to die between initiation of the call [time 1) to follow?up (time 3i).155 Between time 2 {end ofthe call] to time 3, the effect remained for pain and hopelessness, but was not signi?cant for intent to die, suggesting that greater effort at outreach during and following the call is needed for the callers with high levels of suicide intent.155 Treatment for people at risk of suicide. The improving Mood ?Promoting Access to Coiioborative Treatment program aims to prevent suicide among older primary care patients by reducing suicide ideation and depression. iMPACTfacilitates the development of a therapeutic alliance, a personalized treatment plan that includes patient preferences, as well as proactive follow-up {biweekly during an acute phase and during continuation phase} by a depression care manager.156 The program has been shown to signi?cantly improve quality of life, and to reduce functional impairment, depression and suicidal ideation over 24-months of follow-up?swi relative to patients who received care as usual. Coiioborotive Assessment and Management of Suicidoiity (CAMS), is a therapeutic approach for suicide-speci?c assessment and treatment.The program?s flexible approach can be used across treatment settings and clinician theoretical orientations and involves the clinician and patient working together in an interactive assessment process to develop patient-speci?c treatment plans. Sessions are collaborative and involve constant patient input about what is and is not working with the ultimate goal of enhancing the therapeutic alliance and increasing treatment motivation in the suicidal patient. CAMS has been tested and supported in 6 correlational studies,?? in a variety of inpatient and outpatient settings, and in one RCT with several additional RCTs under way. A feasibility trial with a community-based sample of suicidal outpatients randomly assigned to CAMS or enhanced care as usual (intake with a or nurse practitioner followed by 1-11 visits with a case manager and medication as needed] found better treatment retention among the CAMS group and signi?cant improvements in suicidal ideation, overall distress, and feelings of hopelessness at the 12 month follow-up.155 Preventing Suicide: A Technical Package of Fancy. Fmgrammnd Patties 3F Other examples include Dialectical Behavioral Therapy and Attachment?Based Family Therapy a multicomponent therapy for individuals at high risk for suicide and who may struggle with impulsivity and emotional regulation issues.The components of DBT include individual therapy, group skills training, between?session telephone coaching and a therapist consultation team. In a randomized controlled trial of women with recent suicidal or self- injurious behavior, those receiving DBTwere half as likely to make a suicide attempt at the two-year follow-up than women receiving community treatment (23% vs required less hospitalization for suicide ideation, and had lower medical risk across all suicide attempts and self-injurious acts combined?? ABFT is a program for adolescents aged 12?1 8 and is designed to treat clinically diagnosed major depressive disorder, eliminate suicidal ideation, and reduce dispositional anxiety.?5? A randomized controlled trial ofABFTfound that suicidal adolescents assigned to experienced signi?cantly greater improvement in suicidal ideation over 24 weeks of follow-up than did adolescents assigned to enhanced usual care. Additionally, a signi?cantly higher percentage of ABFT participants reported no suicidal ideation in the week prior to assessment at 12 weeks than did adolescents receiving enhanced usual care (69.2% vs. 34.6%} and at 24 weeks [82.1% vs. The Veterans Affairs Translating initiatives for Depression into Effective Solutions project uses a depression care liaison to link primary care and mental health services.The depression care liaison assesses and educates patients and follows-up with both patients and providers between primary care visits to optimize treatment. This collaborative care increases the ef?ciency of providing mental health services by bringing mental health care to the primary care setting, where most patients are ?rst detected and subsequently treated for many mental health conditions. An evaluation of 17055 found signi?cant decreases in depression severity scores among 20% of primary care patients."5 7105? patients also demonstrated 85% and 95% compliance with medication and follow?up visits, respectively.1m Treatment to prevent re attempts. Several strategies that aim to prevent re?attempts have demonstrated impact on reducing suicide deaths. For example, Emergency Department Brief intervention with Follow?up Visits is a program that involves a one?hour discharge information session that addresses suicidal ideation and attempts, distress, risk and protective factors, alternatives to self-harm, and referral options, combined with nine follow?up contacts over 18 months (at 1, 2, 4, 7, 11 weeks and 4, 6, 12, 13 months). Follow-up contacts are either conducted by phone or through home visits according to a speci?c timeline for up to 18 months. A randomized controlled trial that enrolled suicide attempters from eight hospital emergency departments in ?ve countries (Brazil, India, Sri Lanka, lran, and China) found that a brief intervention combined with nine follow-up visits over 18 months was associated with signi?cantly fewer deaths from suicide relative to a treatment-as-usual group versus respectively).162 Another example of treatment to prevent re-attempts involves active follow-up contact approaches such as postcards, letters, and telephone calls intended to increase a patient?s sense of connectedness with health care providers and decrease isolation.151 These approaches include expression of care and support and typically invite patients to reconnect with their provider. Contacts are made periodically or every few months in the ?rst 12 months post?discharge with some programs continuing contact for two or more years}. In a meta?analysis conducted by Inagaki et al?51 interventions to prevent repeat suicide attempts in patients admitted to an emergency department for suicide attempt were found to reduce re-attempts by approximately 12% for up to 12 months post-discharge; however, the effects ofthese approaches beyond 12 months on re-attempts has not yet been demonstrated.151 Also, because the number oftrials and associated sample sizes included in this meta-analysis were small, it was not possible to determine the effect of active contact and follow-up approaches on suicide. 92mm Sul?de: Mahala! mirage of Polity, Preamp. and Practices 6 In a randomized controlled trial of the post?crisis suicide prevention long?term follow?up contact approach, Motto and Eiostrom'f'3 found that patients who refused ongoing care but who were randomized to be contacted by letter four times per year had a lower rate of suicide over two years of follow?up than did patients in the control group who received no further contact. Other studies have also shown post-crisis letters and coping cards to be protective against suicide ideation and Finally, Cognitive Behavior Therapy for Suicide Prevention (CST-SP) is an example of a therapeutic approach to prevent re-attempts. It uses a risk-reduction, relapse prevention approach that includes an analysis of proximal risk factors and stressors relationship problems, school or work-related dif?culties} leading up to and following the suicide attempt; safety plan development; skill building; and CBTASP also has family skill modules focused on family support and communication patterns as well as improving the family?s problem-solving skills. A randomized controlled trial of found that outpatient cognitive therapy designed to prevent repeat suicide attempts resulted in a 50% reduction in the likelihood of a suicide re?atternpt among adults who had been admitted to an emergency department for a suicide attempt relative to treatment as usual.?6 Several strategies that aim to prevent re-attempts have demonstrated impact on reducing suicide deaths. 4 Preventing Suicide: it Technical Package of Policy, Program, and Practices 39 ?7 ?3 .. F. iw'?f by Ger - ya .4 tty Lessen Harms and Prevent Future Risk Rationale Millions of people are bereaved by suicide every year in the United States and throughout the world.5 Risk of suicide and suicide risk factors has been shown to increase among people who have lost a friendfpeer, family member, co-worker, or other close contact to suicide?? Care and attention to the bereaved is therefore of high importance. Despite often good intentions, media and others responding to suicide may add to this risk. For example, research suggests that exposure to sensationalized or otherwise uninformed reporting on suicide may heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to what is known as suicide Approaches Some approaches that can be used to lessen harms and reduce future risk of suicide include postvention and safe reporting and messaging following a suicide. Postvention approaches are implemented after a suicide has taken place and may include debrie?ng sessions, counseling, and/or bereavement support groups for surviving friends, family members, or other close contacts.These programs have not typically been evaluated for their impact on suicide, attempts, or ideation, but they may reduce survivors'guilt, feelings of depression, and complicated grief.?? Safe reporting and messaging about suicide.The manner in which information on a recent suicide is communicated to the public school assemblies, mass media, social media) can heighten the risk of suicide among vulnerable individuals and can inadvertently contribute to suicide contagion. Reports that are inclusive of suicide prevention messages, stories of hope and resilience, risk and protective factors, and links to helping resources hotline], and that avoid sensationalizing events or reducing suicide to one cause, can help reduce the likelihood of suicide contagion.m Potential Outcomes - Reductions in suicidal ideation - Reductions in suicide attempts - Reductions in rates of suicide - Reductions in distress - Improvements in reporting following suicide - Reductions in contagion effects related to suicide Preventing Suicide: ll. Technical Package of Policy, Programs. and Practices 0 Evidence Current evidence suggests that postvention and safe reporting and messaging can impact risk and protective factors for suicide. Postvenliion. One example of a postvention program with evidence of impact on risk and protective factors for suicide is the StondBy Response Service {StondBy}. StondBy provides clients with face-to-face outreach and telephone support through a professional crisis response team. Site coordinators develop customized case management plans, referring clients to other existing community services matched to their needs.?2 In a study by Visser, Comans, and Scufl?ham,?2 StondBy clients were signi?cantly less likely to be at high risk for suicidality (suicide ideation and attempts} and had less distress than a suicide bereaved comparison group who had not had contact with the StondBy program {48% and 64% respectively]. Additionally, research suggests that active postvention approaches in which outreach to suicide survivors occurs at the scene of a suicide is associated with intake into treatment sooner, greater attendance at support group meetings, and attendance at more meetings compared to passive postvention approaches where survivors self-refer for Safe reparting and messaging about suicide- One way to ensure safe reporting and messaging about suicide is to encourage news media to adhere to Recommendations for Reporting on Suicide org). The most compelling evidence supporting these recommendations for reporting comes from Austria. After a sharp increase in suicides on the Viennese subway, media guidelines were introduced and an interrupted time?series design was used to evaluate the national impact of the guidelines on subsequent suicides. Changes in the quality and quantity ofmedia reporting resulted in a nationwide signi?cant reduction of81 suicides annually":19 Finally, research suggests that not only does reporting on suicide in a negative way reporting on suicide and repetition) have harmful effects on suicide, but reporting on positive coping skills in the face of adversity can also demonstrate protective effects against suicide.?4 Reports of individual suicidal ideation (not accompanied by reports of suicide or suicide attempts) along with reports describing a "mastery" of a crisis situation where adversities were overcome was associated with signi?cant decreases in suicide rates in the time period immediately following such reports.?4 Pastven tion and safe reporting and messaging can impact risk and protective factors for suicide. a ?admiral Padraig? "afPallcy, Programs-amt mans 0 Sector Involvement Public health can play an important and unique role in addressing suicide. Public health agencies, which typically place prevention at the forefront of efforts and work to create broad population-level impact, can bring critical leadership and resources to bear on this problem. For example, these agencies can serve as a convener, bringing together partners and stakeholders to plan, prioritize, and coordinate suicide prevention efforts. Public health agencies are also well positioned to collect and disseminate data, implement preventive measures, evaluate programs, and track progress. Although public health can play a leadership role in preventing suicide, the strategies and approaches outlined in this technical package cannot be accomplished by the public health sector alone. As noted in the National Strategy for Suicide Prevention,? the integration and coordination of prevention activities across sectors and settings is critical for expanding the reach and impact of suicide prevention efforts. Other sectors vital to implementing this package include, but are not limited to, education, government (local, state, and federal), social services, health services, business, Iabor,justice, housing, media, and organizations that comprise the civil society sector such as faith-based organizations, youth-serving organizations, foundations, and other non-governmental organizations. Collectively, these sectors can make a difference in preventing suicide by impacting the various contexts and underlying risks that contribute to suicide. The strategies and approaches described in this technical package are summarized in the Appendix along with the relevant sectors that are well positioned to lead implementation efforts. For example, business and labor, the health sector {including insurers, providers, and health systems}, and government entities are in the best position to implement programs and policies that Strengthen Economic Supports and Strengthen Access and Delivery of Suicide Core. These types of supports go beyond individual behavior change and require commitment and support from those sectors that can directly address some of the underlying risks and the environmental contexts that increase the risk for suicide. Public health entities can play an important role by gathering and information to inform policy, raise awareness, and evaluate the effectiveness of various policies. Moreover, partnerships with non-governmental and community organizations can be instrumental in increasing awareness ofand garnering support for policies affecting individuals and families. The public health sector has been at the forefront of many community-based prevention efforts, working collaboratively with schools and community-based organizations, to change social norms and positively impact health behavior. Public health is well suited to take on a similar leadership role in Promoting Connectedness through peer norm and community engagement activities and supporting the development, evaluation, and adoption of effective programs that Teoch Coping and Problem?Solving to prevent the risk of suicide in the ?rst place.These programs are often delivered in school and community settings, making education and non?governmental organizations vital partners in prevention. Businesses, workplaces, and local and state government entities, on the other hand, are in the best position to establish policies and support practices that Create Protective Environments where people live, work, and play. Public health entities can serve in an important role by gathering and information, working with other governmental agencies criminal justice, defense) and agencies within the executive branch of their state or local government in support of policy and other approaches, and evaluating the effectiveness of measures taken. In a similar fashion, public health entities can partner with schools, workplaces, and community organizations to implement and evaluate prevention programs, policies and practices geared toward creating safe, healthy, and supportive environments. Preventing Suicide: A Technical Package of Fancy. Fmgrammnd hitting: 3 Finally, this technical package includes a number of interventions delivered in hospital, primary care, behavioral health care, and community settings designed to identity and SupportPeopie at Risk. The intensity and activities for many of these interventions require the expertise of professionals who are licensed and trained to deliver critical intervention support. The health, social services, and justice sectors can work collaboratively to support individuals at high-risk for suicide and their families. These activities also require coordination of supports across various service providers and community organizations. Regardless of strategy, action by many sectors will be necessary for the successful implementation ofthis package. In this regard, all sectors can play an important and in?uential role in preventing the risk ofsuicide in the ?rst place and lessening the immediate and long?term harms of suicidal behavior by helping those in times ofcrisis get the services and support they need. All sectors can play an important and in?uential role in preventing suicide. Monitoring and Evaluation Monitoring and evaluation are necessary components of the public health approach to prevention. It is important to have timely and reliable data to monitor the extent ofthe problem and to evaluate the impact of prevention efforts. Data are also necessary for prevention planning and implementation. Gathering ongoing and systematic data is important for prevention efforts. However, it is also important to gather data that are uniform and consistent across systems. Consistent I data allow public health and other entities to better gauge the scope of the problem, identify high?risk groups, and monitor the effects of prevention programs and policies. Currently, it is common for different sectors, agencies, and organizations to employ varying de?nitions of suicidal ideation, behavior, and death that can make it dif?cult to consistently monitor speci?c outcomes across sectors and over time. For example, the manner in which deaths are classi?ed can change from one jurisdiction to another, and can change based on local medical and/or medico-Iegal standards.4 CDC's uniform de?nitions and recommended data elements for self-directed violence provide a useful framework to help ensure that data are collected in a consistent manner across surveillance systems.4 Surveillance systems exist at the federal, state, and local levels. It is important to assess the availability of surveillance data and data systems across these levels to identify and address gaps in the systems. CDC's National Vital Statistics System and the National Violent Death Reporting System are examples of surveillance systems that provide data on deaths from suicide. NVSS is a nationwide surveillance system that collects demographic, geographic, and cause-of- death data from death certi?cates? is a state-based surveillance system [currently in 40 states, the District of Columbia, and Puerto Rico) that combines data from death certi?cates, law enforcement reports, and coroner or medical examiner reports to provide detailed information on the circumstances of violent deaths, including suicide, which can assist communities in guiding prevention approaches.?S Data from state and local Child Death Review teams?? and Suicide Death Review Teams (which are in a few states) offer another source to identify deaths and obtain insight into the gaps in services, systems, and modi?able risk factors for suicide. Preventing Suicide: ll Technical Package of Policy, Programs, and Practices 1? The Notional Electronic injury Surveillance System?All injury Program provides nationally representative data about all types and causes of nonfatal injuries treated in US. hospital emergency departments, and can be used to assess national rates of, and trends in, self?harm injuries by cause falls, poisoning, etc), age, racef ethnicity, sex, disposition (where the injured person goes when released from the emergency departmenti.?5 In addition to information on deaths and nonfatal injuries, there are also surveillance systems that provide national, state, and some local estimates of suicidal behavior. The Youth Risk Behavior Surveillance System collects information from a nationally representative sample of 9?1 2 grade students and is a key resource in monitoring health-risk behaviors among youth, including whether youth have seriously considered attempting suicide, attempted suicide, made a plan, or required treatment by a doctor or nurse for a suicide attempt that resulted in an injury, poisoning, or overdose.?7The data are obtained from a national school?based survey conducted by CDC as well as from state, territorial, tribal, and large urban school district surveys conducted by education and health agencies?? The National Survey on Drug Use and Health (NSDUHFG is an annual survey ofthe civilian, non? institutionalized population aged 12 years and older. NSDUH provides both national and state?level estimates of substance use (alcohol, tobacco, illicit drugs, and non-medical use of prescription drugs}; mental health (past year mental illness, co-occurring illnesses); and service utilization, along with suicide ideation, suicide plans, and suicide attempts. NSDUH is a key resource to track trends in suicide-related risk factors in the population and to help identify groups at increased risk?? It is also important at all levels (local, state, and federal} to address gaps in responses, track progress of prevention efforts and evaluate the impact ofthose efforts, including the impact of this technical package. Evaluation data, produced through program implementation and monitoring, is essential to provide information on what does and does not work to reduce rates of suicide and its associated risk and protective factors. Theories of change and logic models that identify short, intermediate, and longvterm outcomes are an important part of program evaluation. The evidence-base for suicide prevention has advanced greatly over the last few decades. However, additional research is needed to understand the impact of programs, policies, and practices on suicide (and suicide attempts, at a minimum), as opposed to merely examining their effectiveness on risk factors. More research is also needed to examine the effectiveness of primary prevention strategies [before risk occurs) and community-level strategies to prevent suicide at the population level. It will be important for researchers to test the effectiveness of combinations of the strategies and approaches included in this package. Most existing evaluations focus on approaches implemented in isolation, but there is potential to understand the synergistic effects within a comprehensive prevention approach. Lastly, there are also many potential opportunities to build and strengthen partnerships across program areas violence prevention, substance abuse prevention) to evaluate the impact of different approaches on multiple outcomes, Preventing Sul?de: Package Polity. Programs, and Practices 0 Conclusion Suicide is a serious public health problem. Rates of suicide have been on the rise for more than a decade and the costs stretch well into the billions of dollars each year. While suicide is a rare outcome statistically, its human impact has a ripple effect that is far-reaching. Each of us likely interacts with suicide survivors, those with lived experience, and those with thoughts of suicide on a daily basis at home, at work, and in our communities. Suicide and suicide attempts are public health issues of societal concern.There are a number of barriers that have impeded progress, including, for example, stigma related to help?seeking, mental illness, being a survivor and fear related to asking someone about suicidal thoughts. Fortunately, like many public health problems, suicide is and more is being done to prevent suicide than ever before, as evidenced by the work ofthe National Action Alliance for Suicide Preventionf'i?fmiBB the release of the ?rst world report on suicide,5 and more timely surveillance data, to namejust a few examples. In an effort to continue pushing the ?eld and society further towards prevention, this technical package includes strategies and approaches that ideally would be used in a comprehensive, multi-level and multi-sectoral way. It includes strategies and approaches to prevent the risk of suicide in the ?rst place, as well as strategies focused on lessening the immediate and long-term harms of suicidal behavior. It includes strategies that range from a focus on the whole population regardless of risk to strategies designed to support people at highest risk. lmportantly, this technical package extends the bounds ofthe typical prevention strategies to consider approaches that go beyond individual behavior change to better address risk factors impacting communities and populations more broadly economic policies to strengthen housing and ?nancial security}. While the evidence base continues to emerge, the collection of programs, policies, and practices laid out here are available for implementation now. In keeping with good public health practice, the intent is that monitoring and evaluation will play a key role in that implementation. Moreover, as new evidence becomes available, this technical package can be re?ned to re?ect the current state of the science. In closing, and in keeping with a message of resilience as spoken by those with lived experience, ?hope, help, and healing is possible." ?Hope, help, and healing is possible.? - ,1 Prevailing Sul?de: A Eadiage'of Policy, Program. and Practices 0 - . References 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. U.S. Of?ce of the Surgeon General, National Action Alliance for Suicide Prevention. 2012 Nationaistrategy forsuicide prevention: goals and objectives for action. Washington, DC: 2012. National Action Alliance for Suicide Prevention. Action Alliance priorities. 201?; http:/i Frieden TH. Six components necessary for effective public health program implementation. Am Public Health. 2014;1 041110242. Crosby AE, Ortega L, Melanson C. Sel?directed violence surveillance: uniform de?nitions and recommended data elements, version 1.0. Atlanta, GA: Centers for Disease Control and Prevention, National Center for injury Prevention and Control; 2011. World Health Organization. Suicide prevention: a global imperative. Geneva, Switzerland: WHO Press; 2014. Centers for Disease Control and Prevention. Web-Based Injury Statistics Query and Reporting System (WISQARS). Atlanta, GA: National Center for Injury Prevention and Control. Available online: Centers for Disease Control and Prevention, National Center for Health Statistics. National Vital Statistics System. KE, Canham-Chervak M, Black SA, Dada E0, Millikan AM, Jones BH. Mental health risk factors for suicides in the US Army, 2007?8. l'nj Prev. 20121861310541 2. Lineberry TW, O'Connor SS. Suicide in the US Army. Mayo Clinic Proceedings. 20128791371078. WL, Spies E, Stone DM, Lokey CN, Trudeau AR, Bartholow B. Suicide rates by occupational group 17 states, 201 2. MM WP Morb Mortal Wkly Rep. 201 ?645. Han B, Crosby AE, Ortega LA, Parks SE, Compton WM, Gfroerer J. Suicidal ideation, suicide attempt, and occupations among employed adults aged 18?64 years in the United States. Compr Kann L, McManus T, Harris WA, et al. Youth Risk Behavior Surveillance - United States, 2015. MMWR CDC SurveiliSumm. Russell ST, Joyner K. Adolescent sexual orientation and suicide risk: evidence from a national study. Am Public Health. 2001 :91 {81:1225?1 281. Stone DM, Luo F, Ouyang L, Lippy C, Hertz MF, Crosby AE. Sexual orientation and suicide ideation, plans, attempts, and medically serious attempts: evidence from local Youth Risk Behavior Surveys, 2001?2009. Am Public Health. Crosby AE, Han B, Ortega LA, Parks SE, Gfroerer J. Suicidal thoughts and behaviors among adults aged e18 years-v United States, 2008-2009. MMWR Lipari R, Piscopo K, Kroutil LA, Kilmer Miller G. Suicidal thoughts and behavior among adults: results from the 2014 National Survey on Drug Use and Health. NSDUH Data Review 2015; 4.pdf. Dahlloerg LL, Krug EG. Violence-a global public health problem. In: Krug EG, Dahlberg LL, Mercy JA, Zwi AB, Lozano A, eds. World Report on Violence and Health. Geneva, Switzerland: World Health Organization; 2002:1a21. Owens D. Fatal and non-fatal repetition of self-harm: systematic review. Br} 20021 8113121934 99. Olfson M, Gerhard T, Huang C, 5, Stroup TS. Premature mortality among adults with schizophrenia in the United States. JAMA 201 5:72(1 21:1 172?1 181. Bossarte HM, Karras E, Lu N, et al. Associations between the Dartment ofVeterans Affairs? suicide prevention campaign and calls to related crisis lines. Public Health Rep (Washington, DC: 1974). Preventing suicide: A technical Package of Pn?cy. Programmed Practices 0? 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 33. 39. 40. 41. 42. Chapman DP, Whit?eld CL, Felitti VJ, Dube SR, Edwards VJ, Anda RF. Adverse childhood experiences and the risk of depressive disorders in adulthood..iAffect Disord. 2004;82l2lz212-225. Dube SR, Anda RF, Felitti VJ, Chapman DP, Williamson DF, Giles WH. Childhood abuse, household dysfunction, and the risk of attempted suicide throughout the life span: ?ndings from the Adverse Childhood Esperiences Study. JAMA. 2001 :286l24l:3089?3096. Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many ofthe leading causes of death in adults: the Adverse Childhood Experiences (ACE) Study. Ami Prev Med. Klomek AB, Sourander A, Gould M. The association of suicide and bullying in childhood to young adulthood: a review of cross-sectional and longitudinal research ?ndings. Can Leeb Lewis T, Zolotor AJ. A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of Lifestyle Medicine. 201 World Health Organization. Global and regional estimates of violence against women: prevalence and health effects of intimate partner violence and non-partner sexual violence. Geneva: World Health Organization; 2013. Bellis MA, Hughes K, Leckenby N, et al. Adverse childhood experiences and associations with health-harming behaviours in young adults: surveys in eight eastern European countries. Bull World Health Organ. 20149291541655. Haegerich TM, Dahlberg LL. Violence as a public health risk. American JournalofLifestyle Medicine. 201 Wilkins N, Tsao B, Hertz M, Davis R, Klevens J. Connecting the dots: an overview of the links among multiple forms of violence. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2014. Hamby S, J.The web of violence: exploring connections among different forms of interpersonal violence and abuse. Briefs in Sociology. New York, NY: Springer; 2013. Kleiman EM, Riskind JH, Schaefer KE, Weingarden H. The moderating role of social support on the relationship between impulsivity and suicide risk. Crisis. Carter M, McGee R, Taylor B, Williams 5. Health outcomes in adolescence: associations with family, friends and school engagement. JAdolesc. Maimon D, Browning CR, Brooks-Gunn J. Collective ef?cacy, family attachment, and urban adolescent suicide attempts. Health Soc Behav. 2010;51l3lz302-324. Capaldi DM, Knoble NB, Shortt JW. Kim HK. A systematic review of risk factors for intimate partner violence. Partner Abuse. Losel F, Farrington DP. Direct protective and buffering protective factors in the development of youth violence. Ami Prev Med. 2012;43l2 Suppl ?58-523. Wyman PA, Brown CH, LoMurray M, et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Public Health. 201 0:1 Dunne EJ, .lL, Dunne?Maxim K, eds. Suicide and its aftermath: understanding and counseling the survivors. New York: Norton; 198?. Mishara BL. The impactofsuicide. New York: Springer; 1995. National Action Alliance for Suicide Prevention: Suicide Attempt Survivors Task Force. The way forwardpathways to hope, recovery, and wellness with insights from lived experience. Washington, DC. Author; 2014. National Action Alliance for Suicide Prevention: Survivors of Suicide Loss Task Force. Responding to grief trauma, and distress afterasuicide: US. national guidelines. Washington, D.C. Author; 2015. Crosby AE, Sacks JJ. Exposure to suicide: incidence and association with suicidal ideation and behavior: United States, 1994. Suicide Life Threat Behav. 2002328132 1 ~328. Cerel J, Maple M, De Venne A, Moore M, Flaherty C, Brown M. Exposure to suicide in the community: prevalence and correlates in one US state. Public Health Rep (Washington, DC: Framing Sumac: Mammal Package of Policy. Progmu. Ind Prawns 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 58. 59. 60. 61. 62. 63. 65. Chapman AL, Dixon-Gordon KL. Emotional antecedents and consequences of deliberate self-harm and suicide attempts. Suicide Life Threat Behav. 2007:32151543-552. Mitchell AM. Kim Y, Prigerson HG, Mortimer?Stephens M. Complicated grief in survivors of suicide. Crisis. 2004:25i1 1:12-18. Sudak H. Maxim K, Carpenter M. Suicide and stigma: a review of the literature and personal re?ections. Acad 2008320121 36?1 42. Cerel J, JL, Neimeyer RA, Maple M, Marshall D. The continuum of ?survivorship": de?nitional issues in the aftermath of suicide. Suicide Life ThreatBehav. Florence C, Simon T, Haegerich T. Luo F, Zhou C. Estimated lifetime medical and work?loss costs of fatal injuries?United States. 2013. MM WR Morb Mortal Wkly Rep. 2015:64138l:1074-1022. Shepard DS, Gurewich D, Lwin AK, Reed GA, Jr, Silverman MM. Suicide and suicidal attempts in the United States: costs and policy implications. Suicide Life Threat Behav. 201 6:46131352?362. Silverman MM, Maris RW. The prevention of suicidal behaviors: an overview. Suicide Life Threat Behav. Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. National Survey on Drug Use and Health. 2016: Desai RA, Dausey DJ, Rosenheck RA. Mental health service delivery and suicide risk: the role of individual patient and facility factors. Am 2005:162i2}:31 1?318. Turecki G. Epigenetics and suicidal behavior research pathways. Am Prev Med. Edwards RW, JumpervThurman P, Plested BA, Getting ER, Swanson L. Community readiness: research to practice. Journal of Community 2000281315291 ~307. Hawkins JD. Catalano RF, Kuklinski MR. Communities that care. ofCriminology and Criminallustice: Springer; 2014:393-408. Plested BA, Edwards RW, Jumper-Thurman P. Community readiness: a handbook forsuccessful change. Fort Collinsr CO: Tri-Ethnic Centerfor Prevention Research. 2006. Luo F. Florence CS. Quispe-Agnoli M. Ouyang L, Crosby AE. impact of business cycles on US suicide rates. 1923?2002. Am Public Health. Fowler KA, Gladden RM, Vagi KJ, Barnes J, Frazier L. increase in suicides associated with home eviction and foreclosure during the US housing crisis: ?ndings from 16 National Violent Death Reporting System States, Public Health. 2015:105l2l:31 1?316. Stack S, Wasserman Economic strain and suicide risk: a qualitative analysis. Suicide Life ThreatBehav. 2007:3711 3:103-1 1 2. Cylus J. Glymour Avendano M. Do generous unemployment bene?t programs reduce suicide rates? A state ?xed? effect analysis covering 1958~2008. Am Epidemiol. Classen TJ, Dunn RA. The effect ofjob loss and unemployment duration on suicide risk in the United States: a new look using mass-layoffs and unemployment duration. Health Econ. 2012:21131338?350. Flavin F: Radcliff B. Public policies and suicide rates in the American states. Social indicators Research. US. Department of Housing and Urban Development. Neighborhood Stabilization Program. hudexchangeJnfo/programs/nspl, 2017. Houle JN, Light MT. The home foreclosure crisis and rising suicide rates, 2005 to 2010. Public Health. Arsenault-Lapierre G, Kim C,Turecki G. diagnoses in 3225 suicides: a meta-analysis. BMC 20043437. Harris EC. Barraclough B. Suicide as an outcome for mental disorders: a meta-analysis. Br} Suicide: Ted'll'ilcal Package ofi?nlicy. programme imam 0 66. 67. 63. 69. 70. 71. 72. 73. 74. 75. 75. 77. 78. 79. 80. 81. 82. 33. 84. BS. 86. 87. 33. 89. Tyrer P, Heed GM, Crawford MJ. Classi?cation, assessment, prevalence, and effect of personality disorder. Lancet. Harris EC, Barraclough B. Excess mortality of mental disorder. 1998;173:1 1?53. Mark TL, Shern DL, Ba galman JE, Cao Z. Ranking America?s mental health: an analysis of depression across the states. Alexandria, VA: Mental Health America. 2007. Lang M.The impact of mental health insurance laws on state suicide rates. Health Econ. Wang PS, Demler D, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Public Health. Cunningham PJ. Beyond parity: primary care physicians? perspectives on access to mental health care. Health Aff (Millwoad). 2009:2813itw490-501. Coffey CE. Building a system of perfect depression care in behavioral health. .lt CommJ QualPatientSaf. 2007533l4l2193-199- The Paul Wellstone and Pete Domenici Mental Health Parity and Addiction Equity Act of 2008 I(lillHPEAi. HR 1424, 110?h Congress. 2000. 0.5. Department of Health and Human Services Health Resources and Services Administrations. Designated health professional shortage areas statistics. 2017. National Action Alliance for Suicide Prevention: Clinical Workforce Preparedness Task Force. Suicide prevention and the clinical workforce: guidelines for training. Washington, D.C.: Author; 2014. Harris KM. Carpenter C, Bao Y.The effects of state parity laws on the use of mental health care. Med Care. Health Resources Services Administration. National Health Service Corps. 2017; US. Department of Health and Human Services Health Resources and Services Administrations. National Health Service Corp clinician retention: a story of dedication and commitment. 2016. Hailey D, Roine R, Ohinmaa A.The effectiveness of telemental health applications: a review. The CanJ 2000:5311 11:769?778. Mohr DC, Vella L, Hart 5, Heckman T, Simon G. The effect of telephone-administered on of depression and attrition: a meta-analysis. Clinical Science and Practice. Coffey CE. Pursuing perfect depression care. Coffey CE, Coffey MJ, Ahmedani BK. An update on perfect depression care. 2013;64t4iz396. Coffey M, Coffey C, Ahmedani BK. Suicide in a health maintenance organization population..lAMA Haddon W. Advances in the epidemiology of injuries as a basis for public policy. Public Health Rep. Toosi M. Labor force projections to 2024: The labor force is growing, but slowly. Washington, D.C.: Bureau of Labor Statistics; Labor Review. 2015:1?33. Noonan ME. Mortality in State Prisons, 2001 -201 4 Bureau ofJustice Statistical Tables. 2016;2501501December}. Knox KL, P?anz S, Talcott GW, et al. The US Air Force suicide prevention program: implications for public health policy. Am Public Health. 2010;1 0011 3124574453, National Action Alliance for Suicide Prevention Workplace Task Force. Comprehensive Blueprintfor Workplace Suicide Prevention. Washington, DC. Author; 2015. Bunyan CW, Becker A, Brandspigel 5, Barber C, Trudeau A, Novins D. Lethal means counseling for parents of youth seeking emergency care for suicidality. WestJf Emerg Med. 2016;17l1 1:3-14. - 51 anlingiuldde: ?technical ncuagentrunq. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 10?. 103. 109. 110. 111. 112. Miller M, Warren M, Hemenway D, Azrael D. Firearms and suicide in US cities. in} Prev. 201 5:21 {e1 }:e1 16-1 19. Crosby AE, Espitia~Hardeman V, Ortega L, Lozano B. Alcohol and suicide. Alcohol: Science, Policy and Public Health. 2013:190?193. Kaplan MS, McFarland BH, Huguet N, et al. Acute alcohol intoxication and suicide: a gender-strati?ed analysis ofthe National Violent Death Reporting System. in] Prev. 201 3:1 9(1 ):38-43. Beautrais AL, Gibb SJ, Fergusson DM, Horwood LJ, Larkin GL. Removing bridge barriers stimulates suicides: an unfortunate natural experiment. Aust NZJ Stokes ML, McCoy KP, Abram KM, GR, Teplin LA. Suicidal ideation and behavior in youth in the juvenile justice system: a review of the literature. Journal of Correctional Health Care. 201 5 :21 Einour AA, Harrison J. Lethality of suicide methods. in} Prev. 2008;1411139?45. Simon 0R, Swann AC, Powell KE, Potter LB, Kresnow MJ, O'Carroll PW. Characteristics of impulsive suicide attempts and attempters. Suicide Life Threat Behav. 2001 :32{1 Deisenhammer EA, CM, Strauss R, Kemmler G, Hinterhuber H, Weiss duration of the suicidal process: how much time is left for intervention between consideration and accomplishment ofa suicide attempt? Clin 2009:?0l1l?9?24. Hawton K. Restricting access to methods of suicide: rationale and evaluation of this approach to suicide prevention. Crisis. 200228611919. Yip P, Caine E, YousufS, Chang S-S, Wu K, Chen Means restriction for suicide prevention. Lancet. 2012:379l9834): 2393-2399. Cox GR, Owens C, Robinson J, et al. Interventions to reduce suicides at suicide hotspots: a systematic review. BMC Public Health. 201 3:1 3(1 Rowhani?Rahbar A, Simonetti JA, Rivara FP. Effectiveness of interventions to promote safe ?rearm storage. Epidemiol Rev. 2016:38l1}:111a124. Hayes Lilli. Suicide prevention in correctional facilities: reflections and next steps. 201336841088? 194. Giesbrecht N, Huguet N, Ogden L, et al. Acute alcohol use among suicide decedents in 14 US states: impacts of off? premise and on?premise alcohol outlet density. Addiction. 2015;1 Escobedo LG, Ortiz M.The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. Xuan Z, NaimiTS, Kaplan MS, et al. Alcohol policies and suicide: a review of the literature. Alcohol Clin Exp Res. 2016;40l10l2043-2055. Cherpitel CJ, Borges GLG, Wilcox HC. Acute alcohol use and suicidal behavior: a review of the literature. Alcoholism: Clinical and EXperirnental Research. 2004286 Pirkis J, Too LS, Spittal MJ, Krysinska K, Robinson J, Cheung YTD. Interventions to reduce suicides at suicide hotspots: a systematic review and meta-analysis. 2015:2[1 11:994?1001. Perron S, Burrows S, Fournier M, Perron PA, Ouellet F. Installation ofa bridge barrier as a suicide prevention strategy in Montreal, Quebec, Canada. Am Public Health. Beautrais AL. Effectiveness of barriers at suicide jumping sites: a case study. 2001 Grossman DC, Mueller BA, Riedy C, et al. Gun storage practices and risk of youth suicide and unintentional ?rearm injuries. JAMA. Mishara BL, Martin N. Effects ofa comprehensive police suicide prevention program. Crisis. 2012:33i3):1 62-168. Knox KL, Litts DA,Ta cott GW, Feig JC, Caine ED. Risk of suicide and related adverse outcomes after exposure to a suicide prevention programme in the US Air Force: cohort study. BMJ. 2003:327i7428):1376. Preventing Suicide: Ted'll'llcal Package oanlicy, programme Practices 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 12?. 128. 129. 130. 131. 132. 133. 134. 135. 136. Hayes LM. Prison Suicide: an overview and a guide to prevention. The Prison Journal. Barker E, Kolves K, De Leo D. Management of suicidal and self-harming behaviors in prisons: systematic literature review of evidence?based activities. Archives ofSuicide Research. 2014:1 881222?240. Rush BR, Gliksman L, Brook R. Alcohol availability, alcohol consumption and alcohol-related damage. 1. The distribution of consumption model..l$tud Alcohol. Gruenewald PJ, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp lies. Lipton H, Gruenewald P. The spatial dynamics of violence and alcohol outlets. Stud Alcohol. 95. Lippy C, DeGue S. Exploring alcohol policy approaches to prevent sexual violence perpetration. Trauma, Violence, ii.- Abuse. Johnson FW, Gruenewald Remer LG. Suicide and alcohol: do outlets play a role? Alcohol Clin Exp Res. Durkheim E. Suicide: a study in sociology (translated by JA Spaulding and Simpson]. New York, NY: Free Press. (Original work published 1892]. 1897/1951. Centers for Disease Control and Prevention. Strategic direction for the prevention ofsuicidal behavior: promoting individual, family, and community connectedness to prevent suicidal behavior. 2009: Available at: Muennig P, Cohen AK, Palmer A, Zhu W.The relationship between ?ve different measures of structural social capital, medical examination outcomes, and mortality. Soc Sci Med. Beyer KM, Layde PM. Hamberger LK, Laud PW. Does neighborhood environment differentiate intimate partner femicides from other femicides? Violence Against Women. Whitley Fl, McKenzie K. Social capital and review of the literature. Harv Rev 2005:1 De Silva MJ, McKenzie K, Harpham T, Huttly SR. Social capital and mental illness: a systematic review. Epidemioi Community Health. 2005:59i812619-622. Wyman PA. Developmental approach to prevent adolescent suicides: research pathways to effective upstream preventive interventions. Am Prev Med. 201-4:478 Suppl 215251?256. Centers for Disease Control and Prevention. Principles of community engagement. Committee on Community Engagement. 1997. Branas CC, Cheney RA, MacDonald JM, Tam VW. Jackson TD, Ten Have TF1. A difference-in-differences analysis of health, safety, and greening vacant urban space. Am.l Branas CC, Kondo MC, Murphy SM, South EC, Polsky D, MacDonald .lM. Urban blight remediation as a cost-bene?cial solution to ?rearm violence. Am Public Health. Bandura A. Social foundations of thought and action: a social cognitive theory. Prentice-Hall, Inc: 1935. Pollock LR, Williams JM. Problem?solving in suicide attempters. 2004:34i112163?162. Herman KC, Borden LA, Reinke WM, Webster-Stratton C. The impact of the Incredible Years parent, child, and teacher training programs on children?s coeoccurring internalizing Knox MS, Burkhart K, Hunter KE. ACT against violence parents raising safe kids program: effects on maltreatment- related parenting behaviors and beliefs. Journal of Family issues. 201 0. Wasserman D, Hoven CW, Wasserman C, et al. School?based suicide prevention programmes: The SEYLE clusterv randomised, controlled trial. Lancet. Wilcox HC, Kellam 56, Brown CH, et aI.The impact of two universal randomized ?rst? and second?grade classroom interventions on young adult suicide ideation and attempts. Drug Alcohol Depend. 2008:95 Suppl 1:560?1?3. Kellam 56, Brown CH, Poduska JM, et al. Effects of a universal classroom behavior management program in ?rst and second grades on young adult behavioral, and social outcomes. Drug Alcohol Depend. 2008:95 Suppl 1255-528. antim] Suidrle: Mahala! Package at Palm. Progmu. and 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. Webster-Stratton C, Jamila Reid M, Stoolmiller M. Preventing conduct problems and improving school readiness: evaluation of the Incredible Years teacher and child training programs in high-risk schools. JournalofChild and Webster-Stratton CH, Reid Beauchaine T. Combining parent and child training for young children with ADHD. Journal of Clinical Child 0 Adolescent 201 91 -203. Reid Mi, Webster-Stratton C. Hammond M. Follow~up of children who received the Incredible Years intervention for oppositional?de?ant disorder: maintenance and prediction of 2?year outcome. Behavior Therapy. 2003:34i4i:471?491. Webster-Stratton C, Hammond M. Treating children with early-onset conduct problems: a comparison of child and parent training interventions. Consult Clin Webster-Stratton C. Reid MJ. Hammond M. Preventing conduct problems. promoting social competence: a parent and teacher training partnership in head start. Clin Child 2001 Spoth RL, Guyll M. Day 5X. Universal family?focused interventions in alcohol?use disorder prevention: cost? effectiveness and cost~bene?t analyses of two interventions. JStud Alcohol. 20026391219228. Curtin SC, Warner M, Hedegaard H. Increase in suicide in the United States, 1999-2014. NCHS Data Brief. Hyattsville, MD: National Center for Health Statistics: 2016. Jobes DA. The Collaborative Assessment and Management of Suicidality (CAMS): an evolving evidence-based clinical approach to suicidal risk. Suicide Life Threat Behav. Wilcox HC, Wyman PA. Suicide prevention strategies for improving population health. Child Adolesc Ciln Am. 201 Isaac M, Elias B, Katz LY, et al. Gatekeeper training as a preventative intervention for suicide: a systematic review. CanJ Gould MS. Munfakh JL, Kleinman M, Lake AM. National suicide prevention lifeline: enhancing mental health care for suicidal individuals and other people in crisis. Suicide Life Threat Behav. 201 2:42i11:22-35. Gilbody S. Bower P. Fletcher J, Richards 0. Sutton AJ. Collaborative care for depression: a cumulative meta?analysis and review of longer?term outcomes. Arch intMed. 2006:166i21iz2314?2321 . ArcherJ, Bower P, Gilbody S, et al. Collaborative care for depression and anxiety problems. Cochrane Database of Systematic Reviews. 2012:10iArt No D0065 25]. Bruce ML, Ten Have TR, Reynolds CF, et al. Reducing suicidal ideation and depressive in depressed older primary care patients: a randomized controlled trial. JAMA. 2004:291 {91:1081-1091. Inagaki M. Kawashima Y, Kawanishi C. et al. Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Affect Disord. 201 5:1 25:66:48. Gould MS, Cross W, Pisani AR, Munfakh JL, Kleinman M. Impact of Applied Suicide Intervention Skills Training on the National Suicide Prevention Lifeline. Suicide Life Threat Behav. 201 Walrath C, Garraza LG, Reid H, Goldston DB. McKeon R. Impact ofthe Garrett Lee Smith youth suicide prevention program on suicide mortality. Am Public Health. 201 310561980993. Godoy Garraza L, Walrath C, Goldston DB, Reid H, McKeon R. Effect of the Garrett Lee Smith Memorial Suicide Prevention Program on suicide attempts among youths.JAMA 2015:72i1 149. Gould MS, Kalafat J, Harrismunfakh JL, Kleinman M. An evaluation of crisis hotline outcomes. Part 2: Suicidal callers. Suicide Life Threat Behav. Hunkeler EM, Katon W, Tang L, et al. Long term outcomes from the IMPACT randomised trial for depressed elderly patients in primary care. BMJ. 2006332175361259-263. Unutzer J, Tang L. Oishi S. et al. Reducing suicidal ideation in depressed older primary care patients. Geriatr Soc. 2006;54l10121550?1556. Preventing Suicide: ii technical ofi'n?tyJ?rogramsamd mam 55F 0 158. 159. 160. 161. 162. 163. 1 64. 165. 156. 157. 168. 169. 170. 171. 172. 173. 1 74. 1?5. 176. 177. Comtois KA, Jobes DA, S. O?Connor 5, et al. Collaborative assessment and management ofsuicidality feasibility trial for next-day appointment services. Depress Anxiety. Linehan MM, Comtois KA, Murray AM, et randomized controlled trial and follow?up ofdialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Arch Gen Diamond GS, Wintersteen MB, Brown GK, et al. Attachment-based family therapy for adolescents with suicidal ideation: a randomized controlled trial. Am Acad Chiid Adolesc 31. Rubenstein Chaney EF, Ober S, et al. Using evidence-based quality improvement methods for translating depression collaborative care research into practice. Families, Systems, 3: Health. 13. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in ?ve countries. Bull World Health Organ. Motto JA, Bostrom AG. A randomized controlled trial of suicide prevention. 2001 :52i6):823- 833. Hassanian-Moghaddam H, Sarjami S, Kolahi AA, Carter GL. Postcards in Persia: randomised controlled trial to reduce suicidal behaviours 12 months after hospital-treated self-poisoning. Br} Wang YC, Hsieh LY, Wang MY, Chou CH, Huang MW, Ko HC. Coping card usage can further reduce suicide reattempt in suicide attempter case management within 3-month intervention. Suicide Life Threat Behav. 2016:46i1):106v120. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, Beck AT. Cognitive therapy for the prevention ofsuicide attempts: a randomized controlled trial. JAMA. Pitman A, Osborn D, King M, Erlangsen A. Effects of suicide bereavement on mental health and suicide risk. Lancet 2014;1z86-94. Etzersdorfer E, Sonneck C1. Preventing suicide by in?uencing mass-media reporting: the Viennese experience, 1980- 1996. Arch Suicide Res. NiederkrotenthalerT, Sonneck G. Assessing the Impact of media guidelines for reporting on suicides In Austria: interrupted time series analysis. Aust 200?;41 (51:419?428. Szumilas M, Kutcher S. Post-suicide intervention programs: a systematic review. CanJ Public Heaith. 201 Bohanna Wang X. Media guidelines forthe responsible reporting of suicide: a review of effectiveness. Crisis. 2012:33i4iz190-193. Visser?v?S, Comans TA, Scuffham PA. Evaluation of the effectiveness of a community-based crisis intervention program for people bereaved by suicide. Journal of Community Cerel J, Campbell FR. Suicide survivors seeking mental health services: a preliminary examination of the role ofan active postvention model. Suicide Life Threat Behav. 2008;38i1 1:30-34. Niederkrotenthaler T, Voracek M, Herberth A, et al. Media and suicide. Papageno Werther effect. BMJ. 2010;341:6841. Centers for Disease Control and Prevention. National Violent Death Reporting System. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2017. Available online: The National Center for the Review 8: Prevention ofChild Deaths. U.S. Child Death Review Programs. Centers for Disease Control and Prevention, Brener ND, Kann L, et al. Methodology ofthe Youth Risk Behavior Surveillance System--2013. MMWR Recomm Rep. - 56 Pum?ngSuidde: sternum: ncuagenrrunq. Pragmatism-aim Appendix: Summary of Strategies and Approaches to Prevent Suicide Best Available Evidence Strategy Practice or Policy Suicide Suicide Attempts or ldeation Other Risk! Protective Factors for Suicide Lead Sectors? Strengthening household ?nancial security Government Strengthen a (local, state, economic hemp oyment ene tprogranis Federal] supports ?3th inmme SUPPOTFS ?l Businessl La bor Housing stabilization policies Gwemmem . (local, state, Nerghborhood Stabilization Program 1/ Federal} Coverage of mental health conditions in health insurance policies Mental Health Parity La ws Government Strengthen Reduce provider shortages in underserved areas (local, state, Federal and National Health Service Corps delivery {If Telemental Health (TMl-li of Healthcare surcide care Safer suicide care through systems change Social Services Henry Ford Perfect Depression Care v" (Pre-cursor to Zero Suicide) Reduce access to lethal means among persons at risk Government intervening at suicide hot spots (local, state) Safe storage practices 1/ Public Health Emergency Department Counseling on Access to lethal Means (so CALM) Healthcare Create Organizational policies and culture Busmessl labor protective Togethe, for We Justice US Air Force Suicide Prevention Program Government . . . . 1/ (local, state, Correctional saicicle prevention Federal) Community-based policies to reduce excessive alcohol use Government (local, state) Alcohol outlet density .f ii" Business/labor Peer norm programs Public Health Promote Sources of Strength Education connectedness Community engagement activities Public Health . Government Greening vacant urban spaces (local) l*This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy, there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing speci?c activities. Preventing Suicide: ATechnical Package of Policy, Program. and Practices Best Available Evidence . . her Risk! n1, Suicide 0t . 1 Strategy Practice or Policy Attempts or Protective Lead Sectors . Factors for Ideatrnn . . Surcrde Social?emotional learning programs Youth Aware ofivlental Health Teach coping :eaEhBPLogramG Education and 00 dirt-or time solving skills Parenting skill and family relationship approaches Public Health The incredible Years Strengthening Families 30? Education Gatekeeper training Applied Suicide intervention Public Health Skills Training Garret Lee Smith Suicide v, Health Care Prevention Program Crisis intervention Public Health National Suicide v? Prevention Lifeline 50'3?! Services Treatment for people at risk of suicide improving Mood Promoting Access v, Identify and to Collaborative Treatment sup port people Collaborative Assessment and Healthcare at risk Management ofSurcrdalrty (CAMS) Social Services Dialectical Behavioral Therapy (DST) v? A ttachment-Based Family Therapy 1/ (A EFT) Translating initiatives for Depression into Effective Solutions project Treatment to prevent re-attempts ED Brief in tervention with v? alth care Follow?up Visrts Active follow?up contact approaches v? Social Services EST for Suicide Prevention Postvention ea care Lessen harms StandBy Response Service and prevent fe . 'd . future risk a reporting an message a out surcr public Health Media Guidelines s/ Media i*This column refers to the lead sectors well positioned to bring leadership and resources to implementation efforts. For each strategy: there are many other sectors such as non-governmental organizations that are instrumental to prevention planning and implementing speci?c activities. Preventing Suicide: A Technical Padrage of Policy, Programs. and "will For more information To learn more about preventing child abuse and neglect, call 1-800-CDC-INFO or visit violence prevention pages at NatiOnal Center for Injury Prevention and Control Division of Violence Prevention 3' Preventing Suicide: A Technical Package of Policy, Programs, and Practices a i na en er or n'ur ven ion am: on re Division of Violence Prevention .. Preventing Suicide: A Technical Package of Policy, Programs, and Practices Developed by: Deb Stone, MSW, MPH Kristin Holland, MPH Brad Bartholow, Alex Crosby, MD, MPH Shane Davis, Natalie Wilkins, 2017 Division of Violence Prevention National Center for Injury Prevention and Control Centers for Disease Control and Prevention Atlanta, Georgia Centers for Disease Control and Prevention Anne Schuchat, MD (RADM, USPHS), Acting Director National Center for injury' Prevention and Control Debra E. Houry, MD, MPH, Director Division of Violence Prevention James A. Mercy, Director Suggested citation: Stone, D.M., Holland, K.M., Bartholow, B., Crosby, A.E., Davis, 5., and Wilkins, N. (2017). Preventing SuiciderA Technical Package ofPoiicies, Programs, and Practices. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention. Preventing Suio'de: A Technical Package of Policy, Programs, and Practices Contents Acknowledgements 5 External Reviewers 5 Overview 7 Strengthen Economic SupportsiS Strengthen Access and Delivery of Suicide Care 19 Create Protective Environments 23 Promote Connectedness 27 Teach COping and Problem-Solving Skills 31 Identify and Support People at Risk 35 Lessen Harms and Prevent Future Risk41 Sector Involvement 43 Monitoring and Evaluation 45 Conclusion 47 References49 Appendix: Summary of Strategies and Approaches to Prevent Suicide 58 Preventing Suicide: I. Technical Package of Policy, Programs, and Practices 3? Acknowledgements We would like to thank the following individuals who contributed in speci?c ways to the development of this technical package. We give special thanks to Linda Dahlberg for her vision, guidance, and support throughout the development of this package. We thank Division, Center, and CDC leadership for their careful review and helpful feedback on earlier iterations of this document. We thank Alida Knuth for her formatting and design expertise. Last but de?nitely not least, we extend our thanks and gratitude to all the external reviewers fortheir helpful feedback, support and encouragement for this resource. External Reviewers Casey Castaldi Prevention Institute Carmen Clelland Of?ce for State, Tribal, Local and Territorial Support, Centers for Disease Control and Prevention Amalia Corby?Edwards American Association Rachel Davis Prevention Institute Pamela End of Horn Indian Health Service Headquarters Craig Fisher American Association Keita Franklin Department of Defense Jill M. Harkavy Friedman American Foundation for Suicide Prevention Jarrod Hindman Colorado Department of Public Health and Environment Linda Langford Education Development Center, Inc. Richard McKeon Substance Abuse and Mental Health Services Administration Doreen 5. Marshall American Foundation for Suicide Prevention Christine Moutier American Foundation for Suicide Prevention Jason H. Padgett Education Development Center, Inc. Jerry Reed Education Development Center, Inc. Dan Reidenberg Suicide Awareness Voices for Education (SAVE) Christine Schuier National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention Morton Silverman Education Development Center, Inc. Ellyson Stout Education Development Center, Inc. Hope M. Tiesman National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention The experts above are listed with their af?liations at the time this document was reviewed. Preventing Suicide: mama Package of Pnlicy. Programs. and Prattires Overview This technical package represents a select group of strategies based on the best available evidence to help communities and states sharpen their focus on prevention activities with the greatest potential to prevent suicide. These strategies include: strengthening economic supports; strengthening access and delivery of suicide care; creating protective environments; promoting connectedness; teaching coping and problem-solving skills; identifying and supporting people at risk; and lessening harms and preventing future risk. The strategies represented in this package include those with a focus on preventing the risk of suicide in the ?rst place as well as approaches to lessen the immediate and long-term harms of suicidal behavior for individuals, families, communities, and society. The strategies in the technical package support the goals and objectives of the Nationai Strategy for Suicide Previewticm1 and the National Action Alliance for Suicide Prevention's priority to strengthen community?based prevention.2 Commitment, cooperation, and leadership from numerous sectors, including public health, education,justice, health care, social services, business, labor, and government can bring about the successful implementation of this package. What is a Technical Package? A technical package is a compilation of a core set of strategies to achieve and sustain substantial reductions in a speci?c risk factor or outcomeFTechnical packages help communities and states prioritize prevention activities based on the best available evidence. This technical package has three components. The ?rst component is the strategy or the preventive direction or actions to achieve the goal of preventing suicide. The second component is the approach. The approach includes the speci?c ways to advance the strategy. This can be accomplished through programs, poiicies, and practices. The evidence for each of the approaches in preventing suicide or its associated risk factors is included as the third component. This package is intended as a resource to guide and inform prevention decision?making in communities and states. Preventing Suicide is a Priority Suicide, as de?ned by the Centers for Disease Control and Prevention (CDC), is part ofa broader class of behavior called sel'f-directed viol'ence. Self-directed violence refers to behavior directed at oneself that deliberately results in injury or the potential for injury? Self-directed violence may be suicidal or non-suicidai in nature. For the purposes of this document, we refer only to behavior where suicide is intended: - Suicide is a death caused by self-directed injurious behavior with any intent to die as a result of the behavior. - Suicide attempt is de?ned as a non?fatal selfndirected and potentially injurious behavior with any intent to die as a result of the behavior. A suicide attempt may or may not result in injury. Suicide is highly prevalent. Suicide presents a major challenge to public health in the United States and worldwide. It contributes to premature death, morbidity, lost productivity, and health care costs.? In 2015 (the most recent year of available death data}, suicide was responsible for 44,193 deaths in the U.S., which is approximately one suicide every 12 minutes?" In 2015, suicide ranked as the 10th leading cause of death and has been among the top 12 leading causes ofdeath since 1975 in the U.S.7 Overall suicide rates increased 28% from 2000 to 2015.5 Suicide is a problem throughout the life span; it is the third leading cause of death for youth 10?14 years of age, the second leading cause of death among people 15v24 and 25234 years of age; the fourth leading cause among people 35 to 44 years of age. the ?fth leading cause among people ages 45?54 and eighth leading cause among people 5434 years of age? Suicide rates vary by racefethnicity, age, and other population characteristics, with the highest rates across the life span occurring among non-Hispanic American Indianz?Alaska Native and non-Hispanic White population groups. in 2015, the rates for these groups were 19.9 and 16.9 per 100,000 population, respectively.?3 Other population groups disproportionately impacted by suicide include middle?aged adults (whose rates increased 35% from 2000 to 2015, with Steep increases seen among both males and females aged 35?64 years?,? Veterans and other military personnel (whose suicide rate nearly doubled from 2003 to 2008, surpassing the rate of suicide among civilians for the ?rst time in decades)?; workers in certain occupational and sexual minority youth, who experience increased suicidal ideation and behavior compared to their non-sexual minority peersu'? Suicides re?ect only a portion of the problem.'5 Substantially more people are hospitalized as a result of nonfatal suicidal behavior suicide attempts) than are fatally injured, and an even greater number are either treated in ambulatory settings emergency departments) or not treated at all.15 For example, during 2014, among adults aged 18 years and older, for every one suicide there were 9 adults treated in hospital emergency departments for self- harm injuries, 27 who reported making a suicide attempt, and over 22?r who reported seriously considering suicideFr?? Suicide is associated with several risk and protective factors. Suicide, like other human behaviors, has no single determining cause. Instead, suicide occurs in response to multiple biological, interpersonal, environmental and societal in?uences that interact with one another, often over time."5 The social ecological model?encompassing multiple levels of focus from the individual, relationship, community, and societal?is a useful framework for viewing and understanding suicide risk and protective factors identi?ed in the literature.? Risk and protective factors for suicide exist at each level. For example, risk factors included:5 - Individual level: history of depression and other mental illnesses, hopelessness, substance abuse, certain health conditions, previous suicide attempt, violence victimization and perpetration, and genetic and biological determinants - Relationship level: high con?ict or violent relationships, sense of isolation and lack of social support, family! loved one's history of suicide, ?nancial and work stress - Community level: inadequate community connectedness, barriers to health care lack of access to providers and medications] - Societal level: availability of lethal means of suicide, unsafe media portrayals of suicide, stigma associated with help-seeking and mental illness. It is important to recognize that the vast majority of individuals who are depressed, attempt suicide, or have other risk factors, do not die by suicide)? Furthermore, the relevance of each risk factor can vary by age, race, gender, sexual orientation, residential geography, and socio-cultural and economic status.LS - Preventing Suicide: A Technical Package of Policy, Program, and Practices Exposure to violence is associated with in creased risk of depression, past-traumatic Protective factors, or those in?uences that buffer against the risk for suicide, can also be found across the different levels of the social ecological model. Protective factors identi?ed in the literature include: effective coping and problem?solving skills, moral objections to suicide, strong and supportive relationships with partners, friends, and family; connectedness to school, community. and other social institutions; availability of quality and ongoing physical and mental health care, and reduced access to lethal These protective factors can either counter a speci?c risk factor or buffer against a number of risks associated with suicide. Suicide is connected to other forms of violence. Exposure to violence child abuse and neglect, bullying, peer violence, dating violence, sexual violence, and intimate partner violence) is associated with increased risk of depression, post-traumatic stress disorder (PTSD), anxiety, suicide, and suicide Women exposed to partner violence are nearly 5 times more likely to attempt suicide as women not exposed to partner violence.26 Exposure to adverse experiences in childhood, such as physical, sexual, emotional abuse and neglect, and living in homes with violence, mental health, substance abuse problems and other instability, is also associated with increased risk for suicide and suicide attempts.2mThe effects of violence in childhood and adolescence can be observed decades later, including severe problems with ?nances, family, jobs, and stress?factors that can increase the risk for suicide. Suicide and other forms of violence often share the same individual, relationship, community, and societal risk factors suggesting that efforts to prevent interpersonal violence may also prove bene?cial in preventing suicide?? CDC has developed technical packages for the different forms of interpersonal violence to help communities identify additional strategies and approaches Further, just as risk factors may be shared across suicide and interpersonal violence, so too may protective factors overlap. For example, connectedness to one?s community,3 school,?- family}3 caring adults??5 and pro-social peers35 can enhance resilience and help reduce risk for suicide and other forms of violence. Preventing Suicide: ll Technical Package of Policy, Programs, and Practices The health and economic consequences of suicide are substantial. Suicide and suicide attempts have far reaching consequences for individuals. families, and communities.?40 In an early study, Crosby and Sacks?H estimated that 7% of the U.S. adult population, or 13.2 million adults, knew someone in the prior 12 months who had died by suicide.They also estimated that for each suicide, 425 adults were exposed, or knew about the death.41 In a more recent study, in one state, Cerel et al?? found that 48% of the population knew at least one person who died by suicide in their lifetime. Research indicates that the impact of knowing someone who died by suicide andfor having lived experience personally have attempted suicide, have had suicidal thoughts, or have been impacted by suicidal loss} is much more extensive than injury and death. People with lived experience may suffer long?term health and mental health consequences ranging from anger, guilt. and physical impairment, depending on the means and severity of the attempt.? Similarly, survivors of a loved one's suicide may experience ongoing pain and suffering including complicated grief,? stigma, depression, anxiety, post- traumatic stress disorder, and increased risk of suicidal ideation and suicide?il46 Less discussed but no less important, are the ?nancial and occupational effects on those left behind.47 The economic toll of suicide on society is immense as well. According to conservative estimates, in 2013, suicide cost $50.8 billion in estimated lifetime medical and work?loss costs alone.?7 Adjusting for potential under?reporting of suicide and drawing upon health expenditures per capita, gross domestic product per capita, and variability among states in per capita health care expenditures and income, another study estimated the total lifetime costs associated with nonfatal injuries and deaths caused by self?directed violence to be approximately $93.5 billion in 201 3.43 The overwhelming burden of these costs were from lost productivity over the life course, with the average cost per suicide being over $1.3 million.43 The true economic costs are likely higher, as neither study included monetary ?gures related to other societal costs such as those associated with the pain and suffering of family members or other impacts. Suicide can be prevented. Like most public health problems, suicide is preventable.? While progress will continue to be made into the future, evidence for numerous programs, practices, and policies currently exists, and many programs are ready to be implemented now. Just as suicide is not caused by a single factor, research suggests that reductions in suicide will not be prevented by any single strategy or Rather, suicide prevention is best achieved by a focus across the individual. relationship, family, community, and societal-levels and across all sectors, private and public.?5 Assessing the Evidence This technical package includes programs, practices, and policies with evidence of impact on suicide or risk or protective factors for suicide. To be considered for inclusion in the technical package, the program, practice, or policy selected had to meet at least one of these criteria: a) meta?analyses or systematic reviews showing impact on suicide; b) evidence from at least one rigorous randomized controlled trial or quasi-experimental design) evaluation study that found signi?cant preventive effects on suicide; cl meta-analyses or systematic reviews showing impact on risk or protective factors for suicide, or d) evidence from at least one rigorous RCT or quasi- experimental design) evaluation study that found signi?cant impacts on risk or protective factors for suicide. Finally, consideration was also given to the likelihood of achieving bene?cial effects on multiple forms of violence; no evidence of harmful effects on speci?c outcomes or with particular subgroups; and feasibility of implementation in a U5. context ifthe program, policy, or practice has been evaluated in another country. Preventing ?technical ?drag! If Ml}. Programim Practise: