Cesarean delivery rates by hospital type among nulliparous and multiparous patients

Abstract

Background Cesarean delivery rates continue to remain high despite recent attempts to decrease these rates. Prior data suggest that there is great variation in cesarean rates by hospital. Objective The intent of this study is to examine the association of several hospital characteristics and cesarean delivery rates in California. Methods We performed a retrospective study of singleton, non-anomalous, term (37–42 week) deliveries in California. We excluded hospitals with <50 deliveries per year and missing hospital information. We separated hospitals by volume based on previously published categories: low-volume (<1200 deliveries/year), medium-volume (1200–2399 deliveries/year), and medium-high-volume (2400–3599 deliveries/year, and high-volume (3600 deliveries/year). We also evaluated rural versus urban and non-teaching versus teaching hospitals. We examined overall cesarean rates as well as stratified by parity and with and without prior cesarean. We analyzed data with chi-square tests and multivariable logistic regression models. Results In a total of 2,545,464 pregnancies, 772,539 (30.35%) resulted in cesarean deliveries. After controlling for race/ethnicity, age, body mass index, education, and insurance, rates of cesarean delivery were higher in low-volume hospitals (aOR: 1.07; 95% CI: 1.0–1.08) and lower in medium-high-volume hospitals (aOR: 0.97; 95% CI: 0.96–0.98) as compared to high-volume hospitals. Rural hospitals had higher rates of cesarean delivery (aOR: 1.08; 95% CI: 1.06–1.10) as compared to urban hospitals while non-teaching hospitals had higher odds of cesarean deliveries (aOR: 1.27; 95% CI: 1.25–1.28) as compared with teaching hospitals. Among nulliparous patients, medium- and medium-high-volume hospitals had lower rates of cesarean deliveries (aOR: 0.95; 95% CI: 0.93–0.96; aOR: 0.93; 95% CI: 0.91–0.94) as compared to high-volume hospitals, while non-teaching hospitals had higher rates of cesarean deliveries than teaching hospitals (aOR: 1.11; 95% CI: 1.10–1.13). Multiparous patients without prior cesarean had higher rates of cesarean delivery at low-volume hospitals and lower rates of cesarean delivery at medium-high-volumes (aOR: 1.07; 95% CI: 1.05–1.10; aOR: 0.96; 95% CI: 0.94–0.098) as compared to high-volume hospitals. Additionally, multiparous patients without prior cesarean had higher rates of cesarean delivery at non-teaching hospitals than teaching hospitals (aOR: 1.16; 95% CI: 1.13–1.19). Multiparous patients with prior cesarean had high rates of cesarean delivery at all volume hospitals with the highest odds at low-volume hospitals (aOR: 1.81; 95% CI: 1.74, 1.89) as well as at rural and non-teaching hospitals. Conclusion Cesarean delivery rates were higher at low and high-volume hospitals for nulliparous and multiparous patients without prior cesarean, but increased with decreasing hospital volume for multiparous patients with prior cesarean. Additionally, cesarean delivery was more likely at rural and non-teaching hospitals. Our results suggest that further investigation is necessary to determine the structural and mechanistic causes of the differences in practice by hospital type in order to identify targets for approaches in reducing cesarean deliveries.