698: Effect of elective induction of labor at 37-40 weeks compared with expectant management on perinatal mortality in California

Abstract

compared with expectant management on perinatal mortality in California Blair Darney, Yvonne Cheng, Jonathan Snowden, Lorie Jacob, James Nicholson, Anjali Kaimal, Sascha Dublin, Darios Gethun, Aaron Caughey Oregon Health & Science University, Obstetrics & Gynecology, Portland, OR, Oregon Health & Science University, Center for Health Systems Effectiveness, Portland, OR, University of California, San Francisco, School of Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences, San Francisco, CA, Group Health Research Institute, Center for Health Studies, Seattle, WA, University of Pennsylvania Health System, Department of Family Medicine and Community Health, Philadelphia, PA, Massachusetts General Hospital, Department of Obstetrics and Gynecology, Boston, MA, Kaiser Permanente Southern California, Department of Research and Evaluation, Pasadena, CA OBJECTIVE: Often, studies of elective induction of labor use spontaneous labor as a comparison group. However, in a randomized trial, induction of labor is compared to expectant management, and thus, increasing gestational age and delivery at a later date. We sought to examine the association between elective induction of labor and expectant management on perinatal death. STUDY DESIGN: We conducted a retrospective cohort study of all singleton, vertex pregnancies at 37-42 weeks in the State of California in 2006. For each gestational age of an elective induction, defined by the Joint Commission criteria, we created a comparison group of women who were undelivered at that gestational age, and who experienced labor at some future gestational age. We excluded chronic conditions, antepartum diagnoses, and fetal anomalies from the expectant comparison group. We compared the outcomes of perinatal death (fetal, neo and post-neonatal deaths) within each group using two group tests of proportions and multivariable logistic regression controlling for common individual and hospital-level confounders. RESULTS: There were 827 perinatal deaths in the cohort (.23%). We found reduced odds of perinatal death among women electively induced at 38, 39, and 40 weeks compared with women expectantly managed at each week (Table). In multivariable analyses, the relationship held at 39 and 40 weeks, but was no longer significant at 38 weeks, possibly due to low numbers of deaths. CONCLUSION: Our findings suggest that elective induction of labor at term, when compared with expectant management, is associated with a reductioninperinatalmortality.Furtherresearchshouldexaminearange of perinatal and maternal outcomes and elective induction of labor at term deserves study in a multicenter, prospective, randomized trial. 699 Hospital rates of cesarean deliveries are negatively associated with severe perineal lacerations in California Blair Darney, Jonathan Snowden, Erika Cottrelll, Jeanne-Marie Guise, Yvonne Cheng, Aaron Caughey Oregon Health & Science University, Obstetrics & Gynecology, Portland, OR, University of California, San Francisco, School of Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences, San Francisco, CA OBJECTIVE: Hospital rates of 3rd and 4th degree lacerations are a quality metric for obstetric care. However, high cesarean rates are also a concern. We explored whether hospital-level rates of cesarean deliveries are associated with rates of 3rd/4th degree lacerations. STUDY DESIGN: We used linked birth registry/hospital discharge data including all deliveries in California in 2006. We excluded all hospitals with 50 deliveries per year. We included 527, 617 deliveries in 268 hospitals. We calculated the proportion of all deliveries at each hospital that were by cesarean and the proportion which resulted in 3rd/ 4th degree lacerations. We used scatterplots, fitted regression lines, and calculated covariances to examine the relationship of cesarean deliveries and 3rd/4th degree lacerations. RESULTS: The proportion of deliveries at hospitals by cesarean is negatively correlated with the proportion of 3rd/4th degree lacerations (covariance -.22) and significantly associated in a multivariable regression model ( -.04, p .0001). The effect is slightly stronger among multiparas than among nulliparas. CONCLUSION: Our findings suggest that hospital-level rates of cesarean deliveries and lacerations are inversely correlated. A focus on reducing lacerations to improve quality measures may increase cesarean rates, which are also of concern.