589 Increases in accidental deaths from prescription opioids among recently pregnant women Margaret Harper 1 , Pedro Luna-Orea 2 , Samuel Lendle 3 1 Mountain Area Health Education Center, Obstetrics and Gynecology, Asheville, North Carolina, 2 North Carolina State Center for Health Statistics, Raleigh, North Carolina, 3 UNC Gillings School of Global Public Health, Biostatistics, Chapel Hill, North Carolina OBJECTIVE: To examine trends in accidental deaths attributed to pre- scription opioids among recently pregnant women in a geographically defined population, the State of North Carolina (NC). STUDY DESIGN: Deaths of pregnant or recently ( 1 year after delivery) pregnant women in NC are identified using three sources: pregnancy- related codes or mention of pregnancy on the death certificate, linkage of electronic death records with livebirth/ fetal death records, and the State hospital discharge database. From the database of these deaths, drug-related deaths occurring between 1996-2007 were reviewed for intent (accidental versus suicide) and by drug category (prescription opioids including methadone versus illicit). Trends in rates of acci- dental deaths due to prescription opioids and illicit drugs were exam- ined across three 4-year periods. RESULTS: A total of 27 accidental deaths from prescription opioids were identified and medical examiner reports were available for 25. Twenty-five (92.6%) were white non Hispanic. The ages ranged from 19-40. Deaths occurred between 7-353 days post delivery. Methadone was mentioned in 18 (66.7%). The number of accidental deaths due to prescription opioids increased 500% from 1996-1999 to 2000-2003 and an additional 233% by 2004-2007. When examined by rates of deaths per 100,000 live births over the three time periods the test of trend was significant for all prescription opioids combined (p0.0001) and for methadone alone (p=0.0004) but not for illicit drugs including heroin and cocaine, p = 0.33. CONCLUSION: Accidental deaths from prescription opioids including methadone are increasing among recently pregnant women. Increases in rates for the general population have paralleled the rise in prescrip- tions for these drugs. Judicious prescribing practices for pain medi- cations for recently pregnant women, as for all patients, should be followed. Accidental Deaths due to Prescription Opioids Including Methadone in N.C. Years No. (rate per 100,000 live births) 1996-1999 1 (0.23) .......................................................................................................................................................................................... 2000-2003 6 (1.27) .......................................................................................................................................................................................... 2004-2007 20 (4.00) .......................................................................................................................................................................................... 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.454 590 Is an abuse history important in predicting the risk of postpartum depression? D. Yvette Lacoursiere 1 , Michael Varner 2 1 University of California, San Diego, San Diego, California, 2 University of Utah, Salt Lake City, Utah OBJECTIVE: Studies reporting that a history of abuse increases the risk of postpartum depression (PPD) have not controlled for depression history or antepartum stressors. This study characterizes the interac- tion of history of abuse, depression and recent maternal stressors on the risk of PPD. STUDY DESIGN: Women delivering a term, singleton, live-born infant were enrolled immediately postpartum. Prepregnancy history of physical or sexual abuse and depression history was identified and coded into a dichotomized exposure variable. Maternal stressors were identified using the “stressor” questions from the Center for Disease Control and Prevention Pregnancy Risk Assessment Monitoring Sys- tem. Women were screened for PPD six to eight weeks postpartum using the Edinburgh Postnatal Depression Scale (EPDS). The primary outcome measure was an EPDS score of 12. RESULTS: 1038 women were included. Psychosocial risk factors were common: abuse history 11.7%, depression history 16.7%, stressors - financial 49.1%, partner-associated 19.8%, emotional 35.0% and traumatic 10.3%. 17.3% screened positive for PPD. More women with an abuse history screened positive for PPD (28.1%) compared to those without (15.4%, p=0.001). In women without pregnancy stres- sors, abuse history did not increase the frequency of PPD (no abuse 9.1%, abuse history 10.0%, p=0.861). While abuse was associated with PPD in a preliminary model (adjusted odds ratio 2.05 [1.28, 3.26], the addition of a depression history, multiple pregnancy stres- sors and both negated the crude association of abuse and PPD (aOR 1.12 [0.66,1.91]). In the full model, prepregnancy BMI, depression history (aOR 2.78 [1.86, 4.16]) and multiple stressors (3 categories aOR 5.71 [2.84, 11.5]; 4 categories (aOR 8.63 [3.01, 24.76]) during pregnancy were associated with screening positive for PPD. CONCLUSION: History of abuse, depression and pregnancy stressors are remarkably common. A depression history and recent maternal stres- sors are strong predictors of screening positive for PPD and may ex- plain the reported association between abuse and PPD. 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.455 591 Low-dose aspirin for prevention of preeclampsia and its complications: a cost-effectiveness analysis Sinae Vogel 1 , Roxanne Rajaii 2 , Geri Ottaviano 1 , Lena Kim 1 , Amanda Yeaton-Massey 1 , Aaron Caughey 1 1 University of California, San Francisco, San Francisco, California, 2 University of California, Berkeley, California OBJECTIVE: Prior studies have demonstrated that low-dose aspirin (ASA) is associated with a reduction in the risk of preeclampsia. We sought to investigate whether such an intervention is cost effective. STUDY DESIGN: A decision analytic model was designed comparing ASA prophylaxis vs. no prophylaxis in a theoretical cohort of 100,000 pregnant women. All costs were derived from the literature, and ex- isting prospective, randomized controlled trials were used to estimate the occurrence of preeclampsia, preterm birth, maternal death, neo- natal death, neurodevelopmental disability, and gastrointestinal bleeding. Utilities were applied to discounted life expectancy to gen- erate QALYs. All costs and QALYs were discounted at 3%. A cost- effectiveness threshold of $100,000 per QALY was utilized. RESULTS: ASA prophylaxis is the dominant strategy as it is on average both cheaper ($18,720 vs. $18,804) and marginally more effective (26.7417 QALYs vs. 26.7422 QALYs). Additionally, negative out- comes associated with pre-eclampsia were significantly reduced in the intervention group, most notably fewer preterm births (10,312 vs 10,251). In sensitivity analysis, when we varied the efficacy of ASA prophylaxis, it remained cost-effective up to a relative risk of 0.91 for all nulliparous women and up to a relative risk reduction of 0.98 for women with a 25% risk of preeclampsia. CONCLUSION: Low-dose aspirin appears to be a cost-effective prophy- laxis for preeclampsia over a wide range of assumptions, particularly for women known to be at elevated risk for preeclampsia. 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.456 Poster Session IV Academic Issues, etc www.AJOG.org S218 American Journal of Obstetrics & Gynecology Supplement to DECEMBER 2009