923: Change in provider culture and cesarean delivery rates: a retrospective cohort study

Abstract

We sought to examine the impact on cesarean delivery (CD) by a change in culture that was mediated through weekly review of all indications of cesarean delivery. We conducted a retrospective cohort study using a natural experiment model that compared two time frames between 2009 and 2013 at a single institution. We introduced a weekly review conference to discuss all cesareans in 2010 and continued that over time. Date of delivery was dichotomized by those delivering before July 1st, 2010 and those delivering after. In a population of 6,424 women with singleton, vertex, term gestations, we examined the rates of CD by date of delivery and results were stratified by parity, BMI, and age. Chi-square tests were employed for statistical comparisons and a p-value of less than 0.05 was used to indicate statistical significance. Our multivariate logistic regression controlled for gestational diabetes, diabetes mellitus, hypertension, preeclampsia, maternal age, and race/ethnicity. Overall, we found that the rate of CD declined significantly pre- and post our intervention in all women (27.4% vs. 22.9%, p<.001) and nulliparous women (30.9% vs. 23.7%, p<.001). The odds ratio of cesarean delivery in all women as related to time cohort is 0.68 (95% confidence interval [CI], 0.59-0.79), meaning that women who delivered after July 1st, 2010 had a 32% reduction in risk of cesarean delivery when compared to women who delivered before that date. Implementation of a weekly review conference of cesarean delivery indications reduced overall cesarean prevalence at our institution. Further research should example specific indications and practice pattern changes. Institution-level efforts to decrease cesarean rates could benefit from such provider conferences.