OKLAHOMA PREGNANCY RISK ASSESSMENT MONITORING SYSTEM VOL 18, NO 1 2014 %HG 6KDULQJ 3UDFWLFHV $PRQJ )LUVW 7LPH 0RWKHUV :KR %UHDVWIHG $W /HDVW (LJKW :HHNV Introduction: The American Academy of Pediatrics (AAP) recommends room sharing (with a crib, bassinet or cradle near the parents bed) for infants, as opposed to bed-sharing (sharing a sleep surface with a parent or other child).1 &XUUHQWO\ QR VSHFL¿F EHG VKDULQJ VLWXDWLRQV DUH FRQsidered safe; epidemiologic studies have shown that bed-sharing is highly associated with an increased risk of sudden infant death syndrome (SIDS), sudden unexpected infant death (SUID) or suffocation.1-2 For the infant, bed-sharing can increase the occurrence of several risk factors for SIDS and SUID, including overheating, airway obstruction, infant head covering, and exposure to tobacco smoke.1 In Oklahoma DPRQJ ¿UVW WLPH EUHDVWIHHGLQJ PRWKHUV): • • • Willinger et al. found that routine bed-sharing in the United States was more likely among mothers younger than 18 years of age, mothers who were African AmerLFDQ RU $VLDQ 3DFL¿F ,VODQGHU PRWKHUV ZLWK DQQXDO household incomes of less than $20,000, those living in southern states, and/or mothers with infants less than 8 weeks of age.2-3 • • Breastfeeding is a protective factor for SIDS.4-6 The protective effect of breastfeeding of any duration increases with exclusivity. A study in Germany found that exclusive breastfeeding reduced the risk for SIDS by half, while breastfeeding of any duration, exclusive or not, has been found to reduce the risk of SIDS when compared to no breastfeeding.3,5,7,8 In addition, breastfeeding has been associated with a lower risk of postneonatal mortality.9 74.7% of mothers reported sharing the bed with their infant at least some of the time. Mothers younger than 30 were more likely than older mothers to report some bed-sharing. Mothers with Medicaid who breastfed for eight or more weeks were more likely to bed-share with their infants (81.5%) compared to mothers without Medicaid (67.5%). African American and American Indian mothers were more likely to bed-share compared to white mothers (33% and 21%, respectively) when controlling for other factors. Prenatal counseling on safe sleep did slightly reduce the risk for bed-sharing, when controlling for confounding factors. Islanders. Hispanics had the lowest proportion. In addition, women who breastfed more than four weeks, had annual family incomes less than $30,000, or were unmarried were more likely to bed-share frequently (always or almost always).10 Another PRAMS study in Breastfeeding has also been associated with an in- Florida found a similar relationship with bed-sharing creased likelihood of bed-sharing. A study, using Ore- and breastfeeding among white and African American gon PRAMS data, reviewed the frequency of bed-shar- mothers who breastfed for four or more weeks.11 ing among racial groups. Results showed that 76.6% of mothers reported to have bed-shared at least some- 7KLV VWXG\ H[SORUHG EHG VKDULQJ SUDFWLFHV DPRQJ ¿UVW times. African Americans had the highest proportion of time mothers who breastfed at least eight weeks postEHG VKDULQJ IROORZHG E\ $VLDQV 3DFL¿F partum in Oklahoma. MATERNAL & CHILD HEALTH SERVICE y OKLAHOMA STATE DEPARTMENT OF HEALTH 1 0HWKRGV Figure 1. Frequency of Bed-Sharing Among First Time Mothers Who Breastfed Eight Weeks or Longer, Oklaho-ma PRAMS 2009-2011 Oklahoma PRAMS survey data for the years 2009-2011 (n=8,834) were used in this report, with an overall unweighted response rate of 68.3%. PRAMS asked new mothers “How often does your new baby sleep or nap on the same sleep surface with you or anyone else?” Mothers could answer “Always, 5 or more times per week but not always, 3-4 times per week, 1-2 times per week, or Never.” For this study, mothers were categorized as “At least sometimes” if they gave any answer RSWLRQ RWKHU WKDQ ³1HYHU ´ 2QO\ ¿UVW WLPH PRWKHUV ZLWK breastfeeding duration of eight weeks or more were analyzed for this study (n=1,120). This was to reduce potential bias from previous breastfeeding and infant sleep experiences. 25.6% 17.4% 23.0% Table 1 highlights the bed-sharing practices (never bedVKDUHG YV VRPHWLPHV RU DOZD\V EHG VKDUHG RI ¿UVW time mothers who breastfed for at least eight weeks by Analysis for this study utilized SAS callable SUDAAN. selected maternal demographics. 3UHYDOHQFH UDWHV DQG FRQ¿GHQFH LQWHUYDO & , HVtimates were calculated. Potential associations were Maternal age, marital status, education, household inLGHQWL¿HG XVLQJ WKH &RFKUDQ 0DQWHO +DHQV]HO &KL FRPH DQG 0HGLFDLG VWDWXV ZHUH VLJQL¿FDQWO\ DVVRFL6TXDUH Ȥ 7HVW 9DULDEOHV ZHUH FRQVLGHUHG VLJQL¿FDQW ated with bed-sharing sometimes or always. Mothers at p <0.05. Multivariate logistic regression models were used to calculate adjusted risk ratios (ARR). Table 1. Bed-sharing Prevalence Among First Time Mothers who Breastfed for Eight Weeks or More by Maternal Demographics, Oklahoma PRAMS 2009-2011 5HVXOWV ,Q 2NODKRPD Q RI ¿UVW WLPH PRWKHUV who breastfed their infants for eight weeks or longer reported that their infant never shared a bed or sleeping surface (Figure 1). However, 17.4% of new mothers reported their infant always bed-shared, and 23.0% bed-shared 5 or more times per week. Almost WKUHH LQ IRXU LQIDQWV Q ERUQ WR ¿UVW WLPH mothers who breastfed eight or more weeks bedshared at least sometimes. Maternal Demographic Age* < 20 years 20-29 years • \HDUV Race White African American American Indian Others Hispanic No Yes Marital Status* Married Other Education*+ < HS HS The Pregnancy Risk Assessment Monitoring System (PRAMS) is an ongoing, population-based study designed to collect information about maternal behaviors and experiences before, during, and after pregnancy. Monthly, PRAMS sampled between 200 to 250 recent mothers taken from the Oklahoma live birth registry. Mothers were mailed up to three questionnaires seeking their participation. Follow-up phone interviews for non-respondents were conGXFWHG $ V\VWHPDWLF VWUDWL¿HG VDPSOLQJ GHVLJQ ZDV XVHG WR \LHOG VDPSOH VL]HV VXI¿FLHQW WR JHQHUDWH SRSXODWLRQ HVWLPDWHV for groups considered at risk for adverse pregnancy outcomes. Information included in the birth registry is used to develop analysis weights that adjust for probability of selection and non-response. > HS Household income* < $25,000 $25,000 - $49,999 $50,000 or more Medicaid* No Yes Ύ ^ŝŐŶŝĮĐĂŶƚ Ăƚ Ɖ фϬ͘Ϭϱ + Excludes mothers less than 18 years old - Cell size less than 30 Never Bed-shared ^ŽŵĞƟŵĞƐ Žƌ ůǁĂLJƐ Bed-shared Rate(%) 95% C.I. Rate (%) 95% C.I. 21.3 22.5 36.7 10.9 - 37.3 17.8 - 28.0 27.6 - 47.0 78.7 77.5 63.3 63.7 - 89.1 72.0 - 82.2 53.0 - 72.4 28.5 - 23.6 - 34.0 71.5 89.1 84.9 78.4 66.0 - 76.4 63.2 - 97.5 66.3 - 94.2 65.1 - 87.6 27.1 16.0 22.6 - 32.1 8.1 - 29.3 72.9 84.0 67.9 - 77.4 70.7 - 91.9 31.3 15.6 26.0 - 37.1 10.0 - 23.4 68.7 84.4 62.9 - 74.0 76.6 - 90.0 16.1 9.5 - 16.0 88.7 83.9 67.9 - 96.7 74.0 - 90.5 30.1 25.0 - 35.8 69.9 64.2 - 75.0 18.6 14.4 40.1 13.1 - 25.8 8.6 - 23.0 32.3 - 48.5 81.4 85.6 59.8 74.2 - 86.9 77.0 - 91.4 51.5 - 67.7 32.5 18.4 26.5 - 39.1 13.3 - 24.9 67.5 81.6 60.8 - 73.5 75.1 - 86.6 MATERNAL & CHILD HEALTH SERVICE y OKLAHOMA STATE DEPARTMENT OF HEALTH 2 Figure 2. Prevalence of Bed-sharing by Infant Safe Sleep Prenatal Care Counseling among First Time Mothers who Breastfed for Eight or More Weeks, Oklahoma PRAMS 2009-2011 less than 20 years of age had the highest bed-sharing rates (78.7%) compared to the other age groups. African American mothers bed-shared more than the other UDFHV 6RFLR HFRQRPLF VWDWXV LQÀXHQFHG EHG VKDULQJ as those mothers with incomes less than $50,000 and those with Medicaid-funded prenatal or delivery care bed-shared more. Rates of bed-sharing ranged from 59.8% for mothers with $50,000 or more annual household income to 89.1% for African American mothers. +RZHYHU WKH FRQ¿GHQFH LQWHUYDOV ZHUH ZLGH LQ PDQ\ cases, indicating small sample sizes. Figure 2 provides the prevalence of bed-sharing practices by the mother’s receipt of prenatal care counseling on infant safe sleep practices. First time, breastfeeding mothers who reported receiving some counseling on safe sleep had a lower prevalence rate of bed-sharing sometimes or always but the difference was not statistically significant in the bivariate analysis. ,QIDQWV RI ¿UVW WLPH PRWKHUV ZKR EUHDVWIHG IRU HLJKW RU more weeks were more likely to bed-share if their mothers had an unintended pregnancy, smoked (before and/or after pregnancy), and reported symptoms of postpartum depresVLRQ 7DEOH (LJKW\ HLJKW SHUFHQW RI ¿UVW WLPH EUHDVWfeeding mothers with postpartum depression symptoms reported infant bed-sharing compared to 73.0% of mothers ZLWKRXW V\PSWRPV 3UHQDWDO FDUH LQ WKH ¿UVW WULPHVWHU DQG Multiple logistic regression analysis was done to PDWHUQDO ERG\ PDVV LQGH[ %0, ZHUH QRW VLJQL¿FDQWO\ DV- DVVHVV WKH IDFWRUV WKDW ZHUH VLJQL¿FDQWO\ DVVRFLDWHG sociated with bed-sharing among this population. ZLWK EHG VKDULQJ SUDFWLFHV DPRQJ ¿UVW WLPH EUHDVWfeeding mothers. Controlling for maternal demoTable 2. Bed-sharing Prevalence Among First Time Moth- graphics, prenatal care, and tobacco use, only three ers who Breastfed for Eight Weeks or More by Selected YDULDEOHV PDLQWDLQHG D VLJQL¿FDQW DVVRFLDWLRQ PDMaternal Experiences, Oklahoma PRAMS 2009-2011 ternal race, prenatal counseling on infant safe sleep, ĞĚͲƐŚĂƌĞĚ ^ŽŵĞƟŵĞƐ and postpartum depression symptoms (Table 3). Maternal Experiences Never Bed-shared Žƌ ůǁĂLJƐ $IULFDQ $PHULFDQ DQG $PHULFDQ ,QGLDQ ¿UVW WLPH Rate (%) 95% C.I. Rate(%) 95% C.I. Prenatal Care in 1st breastfeeding mothers were 33% and 21% more at trimester risk (respectively) to report bed-sharing with their Yes 25.8 21.1 - 31.0 74.2 69.0 - 78.9 infant. Prenatal counseling on infant safe sleep was No 20.9 12.6 - 32.6 79.1 67.4 - 87.4 Pre-pregnancy BMI a protective factor, reducing the risk of bed-sharing, Underweight compared to mothers who did not receive counseling 28.4 22.7 - 35.0 71.6 65.0 - 77.3 Normal on this issue. 19.3 12.6 - 28.6 80.6 71.4 - 87.4 Overweight Obese Pregnancy intention* Intended Unintended Smoking 3 months before pregnancy* No Yes ^ŵŽŬŝŶŐ Ăƚ ƟŵĞ ŽĨ survey* No Yes Postpartum depression symptoms* No Yes 20.4 12.8 - 30.9 79.6 69.1 - 87.2 29.5 18.5 24.3 - 35.4 12.7 - 26.3 70.5 81.5 64.6 - 75.7 73.7 - 87.3 28.2 14.2 23.5 - 33.6 8.2 - 23.5 71.7 85.8 66.4 - 76.5 76.5 - 91.8 27.3 14.8 22.8 - 32.3 7.2 - 28.0 72.7 85.2 67.7 - 77.2 72.0 - 92.8 27.0 12.3 22.5 - 31.9 6.3 - 22.5 73.0 87.7 68.1 - 77.5 77.4 - 93.7 'LVFXVVLRQ Bed-sharing in Oklahoma is a common phenomenon DPRQJ ¿UVW WLPH PRWKHUV ZKR EUHDVWIHHG IRU DW OHDVW HLJKW weeks. Racial differences found in other bed-sharing studies were observed in Oklahoma. Among mothers who breastfed for eight weeks or longer, a protective factor for 6,'6 UDWHV RI EHG VKDULQJ ZHUH VLJQL¿FDQWO\ KLJKHU IRU African American/Black and American Indian mothers. A Missouri study reviewing 119 SIDS and SUID cases found that 47.1% of the infants died while sharing a sleep surface with one or more bed mates. Bed-sharing is associated with infant death, either from SIDS or uninten- Ύ ^ŝŐŶŝĮĐĂŶƚ Ăƚ Ɖ фϬ͘Ϭϱ - Cell size less than 30 MATERNAL & CHILD HEALTH SERVICE y OKLAHOMA STATE DEPARTMENT OF HEALTH 3 Table 3. Adjusted Risk Ratios for Bed-sharing Among First Time Mothers Who Breastfed At Least Eight Weeks, Oklahoma PRAMS 2009-2011 DĂƚĞƌŶĂů ŚĂƌĂĐƚĞƌŝƐƟĐ ĚũƵƐƚĞĚ ZŝƐŬ ZĂƟŽ ; ZZͿΎ One key limitation of this study was the analytic sample size; the sub-analysis produced estimates with wide FRQ¿GHQFH LQWHUYDOV 2WKHU OLPLWDWLRQV IRU WKLV VWXG\ include the lack of measurable information regarding the breadth and depth of prenatal care counseling concerning safe sleep and lack of information about the sleeping environment (including who was bed-sharing with the infant). Social desirability bias may be causing an over-reporting of breastfeeding in the state and/ or an under-reporting of bed-sharing. This study only UHYLHZHG ¿UVW WLPH PRWKHUV ZKR EUHDVWIHG IRU DW OHDVW eight weeks. This group is demographically different from mothers who did not breastfeed for as long or at all (in terms of maternal race, education, age) and may have other distinct health behaviors, other than breastfeeding duration. Causation cannot be assessed, only relationships and associations. 95% C.I. Race White ĨƌŝĐĂŶ ŵĞƌŝĐĂŶ ŵĞƌŝĐĂŶ /ŶĚŝĂŶ Others Ref 1.33 1.21 0.99 1.2 - 1.4 1.1 - 1.4 0.8 - 1.2 Prenatal counseling on infant safe sleep counseling No Yes Ref 0.87 0.8 - .97 Postpartum depression symptoms No Yes Ref 1.16 1.0 - 1.3 5HFRPPHQGDWLRQV Ύ ŽŶƚƌŽůůŝŶŐ ĨŽƌ ŵĂƚĞƌŶĂů ĂŐĞ͕ ĞĚƵĐĂƟŽŶ͕ ŝŶĐŽŵĞ͕ DĞĚŝĐĂŝĚ ƐƚĂƚƵƐ͕ ƉƌĞŶĂƚĂů ĐĂƌĞ͕ ĂŶĚ ƚŽďĂĐĐŽ ƵƐĞ͘ tional asphyxiations and involves unsafe sleep surfac- 1. es.12 The Oklahoma Child Death Review Board found that sleep-related deaths made up over 37% (103 out of 278) of all child deaths reviewed in 2013. Of those 103 cases, 60 (58.3%) were associated with bed-sharing.13 Discuss with all members of the family and other potential caregivers the importance of safe sleep practices, what a safe sleep environment looks like and how to encourage its use, and the importance of roomsharing, not bed-sharing. Mothers who did have prenatal counseling on safe sleep 2. practices for their infant were at a slightly lower risk of bed-sharing compared to mothers who did not receive any counseling on this topic. This reinforces the need for comprehensive counseling on safe sleep at multiple points, during prenatal care, at the delivery hospital and after the infant has come home. However, even 3. among mothers who receive education on infant safe sleep practices, fatigue and lack of quality sleep may override knowledge and usual safety practices. 4. $PRQJ WKH ¿UVW WLPH PRWKHUV ZLWK EUHDVWIHHGLQJ GXUDtion of eight weeks or more, postpartum maternal depression was a slightly elevated risk for bed-sharing. Mothers who were depressed have been found in other research studies to be less likely to follow infant safety guidelines, including infant sleep position recommendations.14 One small study found that infants of mothers who are depressed took longer to fall asleep and had more sleep disturbances, issues that might prompt tired new mothers to bring an infant to bed with them.15 Support and encourage breastfeeding mothers, as every infant feeds and sleeps differently, and reinforce the proximate but separate bed arrangement for optimal breastfeeding and safe sleep success. Support new mothers by allowing them time to sleep without concerns about the infant or household. $GYRFDWH IRU WKH 2I¿FH RI WKH &KLHI 0HGLFDO Examiner and law enforcement agencies to adopt the Centers for Disease Control and Prevention’s PRGHO SROLF\ IRU LQYHVWLJDWLRQ DQG FODVVL¿FDWLRQ of Sudden Unexpected Infant Deaths (SUID) and Sudden Infant Death Syndrome (SIDS), including the use of scene re-creation and digital photography. The methods currently utilized do not adequately provide the opportunity to distinguish accidental overlay (smothering) from undetermined causes and do not collect information available that would identify risk factors. MATERNAL & CHILD HEALTH SERVICE y OKLAHOMA STATE DEPARTMENT OF HEALTH 4 5. 6. 7. 5HIHUHQFHV Provide paid maternity leave for all mothers so they have time to physically recover and bond with WKHLU QHZ LQIDQW ZLWKRXW ORVLQJ ¿QDQFLDO VXSSRUW from their job. Incorporate infant safe sleep messages, including statistics, in all levels of care for the mother/baby dyad, including prenatal care, pediatric visits, and postpartum health visits. Adopt a policy regarding in-house infant safe sleep issues in all Oklahoma birthing hospitals that speci¿HV PRGHOLQJ VDIH VOHHS SRVLWLRQV DQG HQYLURQPHQWV 8. Affordable childbirth classes should be available to all expectant mothers and address safe sleep issues prior to birth. Scholarships should also be available to those who cannot afford classes. 9. Support programs that distribute cribs for low-income families, to include information about the importance of roomsharing and why the crib is safer than an adult bed. 1. American Academy of Pediatrics (AAP). SIDS and Other 6OHHS 5HODWHG ,QIDQW 'HDWKV ([SDQVLRQ RI 5HFRPPHQdations for a Safe Infant Sleeping Environment. PediatULFV 9RO 1R 1RYHPEHU SS H H 2. Willinger M, Hoffman HJ, Hartford RB,(1994). Infant sleep position and risk for sudden infant death synGURPH UHSRUW RI PHHWLQJ KHOG -DQXDU\ DQG National Institutes of Health, Bethesda, MD. Pediatrics ± 3. Willinger M et al (2003). Trends in Infant Bed Sharing in the United States, 1993-2000 The National Infant Sleep 3RVLWLRQ 6WXG\ $UFK 3HGLDWU $GROHVF 0HG 9RO 4. 9HQQHPDQQ 00 %DMDQRZVNL 7 %ULQNPDQQ % HW DO *H6,' Study Group. Does breastfeeding reduce the risk of sudden LQIDQW GHDWK V\QGURPH" 3HGLDWULFV 9RO 1R 0DUFK 2009. 5. Hauck FR, Thompson (2011) JM, Tanabe KO, Moon RY, 9HQQHPDQQ 00 %UHDVWIHHGLQJ DQG UHGXFHG ULVN RI VXGGHQ LQIDQW GHDWK V\QGURPH D PHWD DQDO\VLV 3HGLDWULFV ± 6. Zotter (2011) Breast feeding is associated with decreased risk of sudden infant death syndrome. Pediatrics ± 10. Support the Infant Safe Sleep workgroup of OSDH’s Preparing for a Lifetime, It’s Everyone’s Responsibility’s initiative in its educational efforts and promotion 7. of current American Academy of Pediatrics safe sleep recommendations. 8. Ip S, Chung M, Raman G, Trikalinos TA, Lau J. A summary of the Agency for Healthcare Research and Quality’s evidence report on breastfeeding in developed FRXQWULHV %UHDVWIHHG 0HG VXSSO 6 ±6 9. Chen A, Rogan WJ. Breastfeeding and the risk of postneonatal death in the United States. Pediatrics. 2004;113 (5). 11. Strengthen the referral network for mothers with symptoms of postpartum depression, as some areas do not have adequate services for new mothers. 12. Discuss and assess, via focus groups, those barriers that relate to safe sleep to gain an understanding of why new mothers bed-share, even if they have received education and have resources for providing safe sleep environments. Moon R. and Fu L (2012). An Update on Sudden Infant Death Syndrome; Pediatrics in Review 2012;33;314, '2, SLU 10. Lahr MB, Rosenberg KD, Lapidus JA. Maternal-Infant %HG VKDULQJ 5LVN )DFWRUV IRU %HG VKDULQJ LQ D 3RSXlation-Based Survey of New Mothers and Implications for SIDS Risk Reduction. Matern Child Health J. 2007 0D\ 13. Understand that there are many cultural perspectives associated with bed-sharing, and that educational DSSURDFKHV FDQQRW EH ³RQH VL]H ¿WV DOO´ LI WKH\ DUH WR be successful. %URXVVDUG '/ 6DSSHQ¿HOG :0 *RRGPDQ '$ 7KH Black and White of infant back sleeping and infant bed sharing in Florida, 2004-2005. Matern Child Health - $SU GRL V 0768-y. 14. Refer all families to the Tobacco Quitline (1-800-Quit-Now) if someone in the household smokes. MATERNAL & CHILD HEALTH SERVICE y OKLAHOMA STATE DEPARTMENT OF HEALTH 5 12. Kemp JS, et al. Unsafe Sleep Practices and an Analy14. Paulson JF, Dauber S, Leiferman JA. Individual and sis of Bedsharing Among Infants dying Suddenly and Combined Effects of Postpartum Depression in Mothers 8QH[SHFWHGO\ 5HVXOWV RI D )RXU