Suicides
Involving Veterans

Arizona Violent Death
Reporting System
January 1, 2015 – December 31, 2017

 5,362 (93.9%) of the decedents. From these, we
excluded 1,046 (19.5%) homicides and 638 (11.9%)
violent deaths of undetermined manner, leaving 3,678
(68.6%) suicides for analysis. We further excluded 77
(2.1%) cases for which decedents’ veteran status was
unknown, after which our sample consisted of 3,601
The Arizona Violent Death Reporting System
(AZ-VDRS) collects violent death data from multiple

suicides for which circumstance and veteran status
data were available.

sources: death certificates issued by the Arizona

We determined veteran status using the indicator for

Department of Health Services (ADHS), police reports

military veteran on the official death certificate; we

obtained from investigating agencies, and autopsy

did not seek external validation, and thus, our data

reports from medical examiner offices. The purpose

may overcount non-veterans as veterans. Use of this

of this project is to assist stakeholders with strategic

definition is consistent with NVDRS standards and

planning and prevention efforts aimed towards

with prior research.1 Note that the term veteran may be

reducing the number of violent deaths that occur each

defined differently elsewhere; for example, individuals

year in Arizona. The data used for this report – Suicides

who are ineligible for benefits based on discharge

Involving Veterans – were drawn from the compilation

status may be excluded in other contexts. AZ-VDRS

and analysis of three years of AZ-VDRS data, from

data analyses and rate calculations also may differ from

January 1, 2015 through December 31, 2017.

those of other sources such as the ADHS when our

AZ-VDRS recorded a total of 5,711 violent deaths for

respective analytic processes differ.

this period; circumstance data were available for

EXHIBIT 1:
PERCENTAGE OF SUICIDES BY VETERAN STATUS, 2015-2017 (N=3601)
■ 

21.1%

53.2%
● Non-Veteran
● Veteran

During 2015–2017, in Arizona,
veterans comprised more than 1
in 5 (21.1%) of all suicide victims.

 For example, AZ-VDRS counts occurrent deaths (those occurring within
the state, regardless of legal residency) rather than resident deaths (those
The

of Arizona residents, regardless of location where death occurs). AZ-VDRS

purpose

analyses include only decedents for whom we have sufficient data rfom ht e
sources noted above, including but not limited to official death certificates.
As a result, AZ-VDRS and ADHS reports overlap; at the same time, these
organizations can each offer unique insights reflecting ht eir er spective analytic

of this project is
to assist stakeholders
with strategic

planning and
prevention

strategies. For this report, there are no known systematic errors in the
AZ-VDRS veteran status counts.

efforts aimed towards

According to a report by the US Department of Veterans Affairs, veterans were
about 41% more likely to be at risk of suicide mortality than the general US
population . The US Census estimated that veterans accounted for more than
11% of Arizona’s population, which was significantly higher than the national
estimate of about 9% (December 2015). Taken together, veteran suicides

reducing
the number of

violent deaths
that occur each year in

represent a substantial problem for the nation, and Arizona in particular.

Arizona

For population estimates, we relied on the American Community Survey
(US Census) 5-year estimates for 2015, 2016, and 2017 to compute crude rates where rates are presented. Note
that in all of the exhibits below, data and analyses represented are for the state of Arizona, 2015–2017, unless
otherwise indicated.

This AZ-VDRS briefing is offered to assist with suicide prevention in Arizona. We also want to underscore
the final words from the National Veteran Suicide Prevention Annual Report 20192:

Suicide is a national issue, with rising rates of suicide in the
general population. In addition, suicide rates are higher, and are
rising faster, among veterans than among non-veteran adults.
Every death by suicide is a tragedy that affects individuals and communities.
Unfortunately, no one strategy in isolation has been shown to be effective in
ending suicide. We must come together to address systematically the larger
societal issues fueling the increased rates of suicide in our nation, keeping at
the forefront of our minds that we

prevent suicide through

meaningful connection, one person at a time.

 EXHIBIT 2:
SUICIDE RATES PER 100,000 POPULATION BY SEX* AND VETERAN STATUS,
2015-2017 (N=3601)

60
Rate per 100,000 Population

53.1
50
40
32.9
28.9

30
20
10.7

10
0

Male

● Non-Veteran

■

Overall suicide rates per
100,000 population were
significantly higher for male
victims, than for female victims
(36.6, 11.0, (not shown).

■

Female

● Veteran

Males who were veterans were
at significantly greater risk for
committing suicide than males
who were not veterans; during
this period, the suicide rate for
veterans was 61.4% greater
than the rate for their nonveteran counterparts (53.1,
32.9).

■ 

Female veterans were almost
three times more likely to commit
suicide than females who were
not veterans (28.9, 10.7).

 EXHIBIT 3:
SUICIDE RATES PER 100,000 POPULATION BY RACE/ETHNICITY* AND VETERAN
STATUS, 2015–2017 (N=3601)

57.8

Rate per 100,000 Population

60
50
40
30

25.8
21.8

20

24.0
19.7

19.8

14.8
10.8

10

11.6
5.4

0
● Non-Veteran

● Veteran

*  Statistically significant at p ≤ .05
†
 Non-Hispanic/Latino
††
 Includes Asian, Native Hawaiian, Pacific Islander, Other and Unspecified

■ 

Across racial/ethnic groups, relative suicide rates for
veterans and non-veterans differed significantly.

■ 

The suicide risk was highest for white non-Hispanic
veterans, with a rate of 57.8 per 100,000 population.

■ 

Within most racial/ethnic groups, veterans were
at greater risk of suicide than non-veterans; the
exception was Native Americans, for whom the
suicide rate for non-veterans was almost double the
rate of veterans (19.7, 10.8).

 EXHIBIT 4:
SUICIDE RATES PER 100,000 POPULATION BY AGE GROUP* AND VETERAN STATUS,
2015-2017 (N=3601)

100

Rate per 100,000 Population

90
80

78.4

70

62.1

60
50

40.0

40
30

43.4

18.9

20.7

18-34

35-54

46.9

24.0

20

16.2

15.5

64-74

75 or older

0

55-64

70
-7
4

10

Age Group (Years)
● Non-Veteran
● Veteran
* Statistically significant at p ≤ .05

■ 

Across all age groups, non-veteran suicide rates
remained relatively level, ranging from 15.5 for those
75 and older to a high of 24.0 for those ages 55–64;
regardless of age group, the rate for non-veterans
was never higher than that for veterans.

■ 

Across all age groups, veterans ages 18–34 had the
highest suicide rate (78.4); this rate was lower for
those between ages 35–54 (40.0), then gradually
increased with age to 62.1 for those 75 or older.

 EXHIBIT 5:
SUICIDE RATES PER 100,000 POPULATION BY COUNTY* AND VETERAN STATUS,
2015-2017 (N=3601)

90.0
79.3

Rate per 100,000 Population

80.0

74.7

70.0

62.1

60.0
50.0
41.1

40.0

42.4

44.1

46.6

30.0

47.9

48.7 50.4

51.1

52.8

51.7

29.3

26.3
21.8
17.8
14.7 13.7

20.0

8.0

10.0

19.7
11.1

55.3

18.8

21.3

20.5

11.3

32.8 33.0 33.8

16.2
9.3

z

Ap ila
ac
he
P
G inal
re
en
l
C ee
oc
h
M ise
ar
i
A cop
R
IZ a
O
N
A
Pi
m
a
Yu
m
La a
Pa
N z
av
a
Ya jo
va
C pai
oc
on
i
M no
oh
av
e

G

ru
C

a
nt

Sa

G

ra

ha

m

0.0

County
● Non-Veteran

● Veteran

* Statistically significant at p ≤ .05
■ 

In Arizona, during 2015-2017, the statewide suicide
rate among veterans was almost twice that of
non-veterans (50.4, 21.3 per 100,000 population).

■ 

Suicide rates for veterans were substantially and
significantly higher than rates for non-veterans in
every Arizona county.

■ 

In Santa Cruz County, the suicide rates for veterans
and non-veterans were most similar, at 17.8 and 13.7,
respectively.

■ 

Mohave County (79.3) had the highest veteran
suicide rate, followed closely by Coconino County
(74.7); Graham and Santa Cruz had the lowest rates
(14.7, 17.8).

■ 

Apache, Greenlee, and La Paz counties each had
a greater than 3-to-1 ratio of veteran to non-veteran
suicide rates.

ARIZONA
COUNTIES

 EXHIBIT 6:
EDUCATION COMPLETED, MARITAL STATUS, AND BIRTHPLACE AMONG SUICIDE
VICTIMS AGES 18 AND OLDER BY VETERAN STATUS, 2015-2017 (N=3483)

NON-VETERAN

VETERAN

TOTAL

n

%

n

%

N

%

<= 8th grade

80

2.9

11

1.4

91

2.6

9th – 12th grade

315

11.6

35

4.6

350

10.0

High school grad or GED

973

35.7

256

33.6

1229

35.3

Some college credit

553

20.3

192

25.2

745

21.4

Associate or bachelor’s degree

570

20.9

183

24.0

753

21.6

Advanced degree

174

6.4

73

9.6

247

7.1

Unknown

57

2.1

11

1.4

68

2.0

Never married

1048

38.5

118

15.5

1166

33.5

Married

723

26.6

287

37.7

1010

29.0

Married, but separated

134

4.9

28

3.7

162

4.7

Divorced

654

24.0

228

30.0

882

25.3

Widowed

131

4.8

96

12.6

227

6.5

Single, unspecified

7

0.3

0

0.0

7

0.2

Unknown

25

0.9

<5

na

25

0.8

Arizona

728

26.7

103

13.5

831

23.9

Other US state or territory

1679

61.7

629

82.7

2308

66.3

Foreign country

268

9.8

18

2.4

286

8.2

47

1.7

11

1.4

58

1.7

Education Completed*

Marital Status*

Birthplace*

Unknown
* Statistically significant at p ≤ .05

Note: CDC reporting requirements require that counts less than 5 not be shown for reasons related to data reliability and identity protection.
These counts can, however, be included in totals. Therefore, totals in each row may include values represented here only as <5.

 Veteran suicide victims differed significantly from
non-veteran victims with respect to
education completed, marital status, and birthplace.

Veteran suicide victims
were substantially
more likely to have earned
some college credit or
a degree, compared
to non-veterans
(58.8%, 47.6%).

Veteran suicide victims
were also significantly
more likely than non-veteran
victims to have been married,
including married but
separated (41.4%, 31.5%),
or divorced (30.0%, 24.0%).

Non-veteran suicide
victims were more than
twice as likely as veteran
victims to have
never married
(38.5%, 15.5%).

Veteran suicide victims
were significantly more
likely than non-veteran
victims to have been
born in a US state
other than Arizona
(82.7%, 61.7%).

 EXHIBIT 7:
LOCATIONS OF SUICIDE BY VETERAN STATUS, 2015-2017 (N=3601)

NON-VETERAN

VETERAN

TOTAL

n

%

n

%

N

%

2047

72.1

595

78.2

2642

73.4

Street/road, sidewalk, alley

118

4.2

26

3.4

144

4.0

Motor vehicle (excluding school bus,
and public transportation)

150

5.3

33

4.3

183

5.1

Commercial establishment
(e.g., bar, store, service station)

26

0.9

<5

na

26

0.8

Parking lot/public parking garage

58

2.0

21

2.8

79

2.2

Jail, prison, group home, shelter,
other supervised residential facility

57

2.0

<5

na

57

1.6

Park, playground, public use area

40

1.4

14

1.8

54

1.5

Natural area
(e.g., field, river, beach, woods)

145

5.1

31

4.1

176

4.9

Hotel/motel

88

3.1

20

2.6

108

3.0

Other

93

3.3

14

1.8

107

3.0

Unknown

18

0.6

<5

na

18

0.5

Location*
House or apartment

* Statistically significant at p ≤ .05
Note: CDC reporting requirements require that counts less than 5 not be shown for reasons related to data reliability and identity protection.
These counts can, however, be included in totals. Therefore, totals in each row may include values represented here only as <5.

■ 

Among both veteran and nonveteran suicide victims, about
3 in 4 suicides were committed
in private residences.

■ 

Although locations where
suicides occurred varied
significantly among veteran
and non-veteran victims, for any
single location type, there were
few substantive differences
between them.

■

Notably, less than 0.5%
(n<5) committed suicide
while in jail, prison, shelter, or
other supervised facility,
compared to about 2.0% of
non-veteran suicide victims.

 EXHIBIT 8:
METHODS OF DEATH BY VETERAN STATUS, 2015-2017 (N=3601)

NON-VETERAN

VETERAN

TOTAL

n

%

n

%

N

%

1517

53.4

609

80.0

2126

59.0

Sharp instrument

41

1.4

10

1.3

51

1.4

Blunt instrument

95

3.3

5

0.7

100

2.8

Hanging, strangulation, suffocation

713

25.1

74

9.7

787

21.9

Poisoning

441

15.5

59

7.8

500

13.9

Othera

33

1.2

<5

na

33

1.0

Unknown

0

0.0

0

0.0

0

0.0

Methods*
Firearm

*  Statistically significant at p ≤ .05
a
  Including, but not limited to falls, fire/burns, motor vehicles and drowning.
Note: CDC reporting requirements require that counts less than 5 not be shown for reasons related to data reliability and identity protection.
These counts can, however, be included in totals. Therefore, totals in each row may include values represented here only as <5.

■ 

Between veteran and nonveteran suicide victims, there
were significant differences in
the methods or causes of death.

■ 

Notably, most veteran suicide
victims used a firearm, compared
to about half of non-veteran
victims (80.0%, 53.4%).

■ 

Veteran suicide victims also
used hanging, strangulation, or
suffocation (9.7%) or poisoning
(7.8%) far less frequently than
non-veteran suicide victims
(25.1%, 15.5%, respectively).

 EXHIBIT 9:
CIRCUMSTANCES OF SUICIDE VICTIMS BY VETERAN STATUS, 2015–2017 (N=3601)
NON-VETERAN
n

%

VETERAN
n

%

TOTAL
N

%

Mental Health Issues
Current mental health problem*

1299

47.5

312

33.2

1611

43.8

Current depressed mood*

1106

40.4

262

27.8

1368

37.2

Ever treated for mental illness or substance misuse*

861

31.5

145

15.4

1006

27.4

Current treatment for mental illness or substance misuse*

628

22.9

113

12.0

741

20.1

Total victims w/ one or more mental health factors*

1824

66.6

441

46.9

2265

61.6

Alcohol problem*

520

19.0

117

12.4

637

17.3

Other substance problem*

582

21.3

59

6.3

641

17.4

Substance Abuse / Addiction

Other addiction (gambling, sexual, etc.)

21

0.8

<5

na

24

0.7

939

34.3

155

16.5

1094

29.7

Family relationship problem

291

10.6

39

4.1

330

9.0

Intimate partner problem*

766

28.0

129

13.7

895

24.3

Other relationship problem

61

2.2

8

0.9

69

1.9

Total victims w/ one or more addiction factors*

Interpersonal Issues

Perpetrator of interpersonal violence in past month*

71

2.6

17

1.8

88

2.4

Victim of interpersonal violence in past month*

16

0.6

<5

na

17

0.5

Suicide of friend/family in past 5 years

54

2.0

12

1.3

66

1.8

Other death of friend/family*

151

5.5

51

5.4

202

5.5

Total victims w/ one or more interpersonal factors*

1184

43.3

222

23.6

1406

38.2

Life Stressors
Physical health problem*

606

22.1

321

34.1

927

25.2

Job problem*

308

11.3

41

4.4

349

9.5

Recent criminal related legal problem*

209

7.6

41

4.4

250

6.8

Other legal problems*

67

2.4

10

1.1

77

2.1

Financial problem*

289

10.6

49

5.2

338

9.2

School problem*

41

1.5

<5

na

44

1.2

Eviction or loss of home

112

4.1

23

2.4

135

3.7

1275

46.6

421

44.7

1696

46.1

Previous attempts*

706

25.8

120

12.8

826

22.5

Disclosed intent to commit*

858

31.3

234

24.9

1092

29.7

Suicidal thoughts*

1338

48.9

339

36.0

1677

45.6

Total victims w/ one or more historical factors*

1723

63.0

417

44.3

2140

58.2

Total victims w/ one or more life stressor factors*

Suicidal History

* Statistically significant at p ≤ .05
Note: Circumstance characteristics are not mutually exclusive, and any particular victim may have any number of circumstances present.

 Selected Circumstance Findings

■ 

Veteran suicide victims were less likely than non-veteran victims
to have mental health and/or substance misuse issues reported;
for example, one or more mental health-related circumstances
were reported for 46.9% of veteran victims compared with
66.6% of non-veteran victims.

■ 

Substance misuse problems, not including alcohol, were
reported more than three times as often for non-veteran suicide
victims as for veteran victims (21.3%, 6.3%).

■ 

Interpersonal problems appeared to be a less significant factor
for veteran suicide victims than for non-veteran victims; some
form of interpersonal problem was reported for a little more than
1 in 5 veteran victims, and just over 2 in 5 non-veteran victims
(23.6%, 43.3%)

■ 

Conversely, a physical health problem was more likely to have
been reported for veteran suicide victims than for non-veteran
victims (34.1%, 22.1%).

■

Suicide victims who were veterans were significantly less
likely than non-veteran victims to be reported as having a
history of attempting suicide (12.8%, 25.8%); in fact, veteran
victims were less likely to have any prior indicators of suicide
risk reported (44.3%, 63.0%, respectively).

 Implications/Recommendations on
Suicides Involving Military Veterans
Suicide among military veterans is a critical and

Most veteran suicide victims in our analyses were

emerging issue nationally, and this is of paramount

male. It may be a lingering cultural influence that men

concern in the State of Arizona, where AZ-VDRS

generally and veterans specifically are disinclined

findings show a significant and substantial influence

to reach out for help when experiencing mental and

of veteran status on the individual suicide risk. The

emotional distress; this suggests that early screening

proportion of veterans in the state population is higher

and treatment of both male and female veterans with

than the national average. Given the geographic size

risk factors for depression is particularly important

and rural nature of much of the state, dispersion of

for suicide prevention. More than 33% of all veteran

resources becomes a critical component of responding

suicide victims (not only males) in this report had been

to veteran suicides.

diagnosed with depression or dysthymia (depressed

Our analyses showed that suicide victims who
were veterans were less often reported to have had
substance abuse and interpersonal problems or
conflicts than non-veteran victims. Veteran and nonveteran victims were similar in their associations with
life stressors in general, but veteran victims were more
likely to have had serious physical health problems that
may have contributed to the suicide—for as many as 1 in
3 veteran victims, this may have been the suicide trigger.
This fact suggests that veterans with major physical
health issues are in need of far more immediate and
effective support throughout even prolonged periods
of dealing with the physical and emotional trauma and
challenges presented by physical impairment.

mood) prior to taking their own lives, yet only 12%
were currently receiving treatment. Further, more than
a third were known to have had suicidal thoughts and
a quarter had disclosed their intent to commit suicide
shortly before doing so. If we as a state and a nation are
serious about preventing suicide among our veterans,
increased support for mental health screening and
treatment after diagnosis is needed urgently. Critically,
we owe veteran men and women the highest standard
of care and a rapid, effective response when they have
disclosed suicidal thoughts and intentions or have
survived actual attempts. The goal should be nothing
less than the restoration of their potential for quality
of life.

END NOTES
  Huguet, N., Kaplan, M. S., & McFarland, B. H. (2014). The effects of misclassification biases on veteran suicide rate estimates. American journal
of public health, 104(1), 151-155.

1

2

National Veteran Suicide Prevention Annual Report 2019. (September 2019). US Dept. of Veterans Affairs, Office of Mental Health and
Suicide Prevention.

3

AZ-VDRS estimates of suicide rates, particularly those of Native American males, may differ from rates reported by other death surveillance
systems, due to important variations in data sources and coding protocols. For this reason, comparative analyses outside NVDRS and
AZ-VDRS should be approached with caution.

cvpcs.asu.edu