Maternal Clinical Diagnoses and Hospital Variation in the Risk of Cesarean Delivery

Abstract

Cesarean delivery (CD) is the most common inpatient surgery in the United States. The risks of CD-related maternal and neonatal complications depend on clinical conditions, but hospital-level differences in the use of CDs also vary widely. A better understanding of variations in CD use may improve consistency, quality, and value for parturients and their infants. In this retrospective multilevel analysis, hospital discharge records were used to determine the extent to which variability in the risk of CD in US hospitals was attributable to maternal clinical diagnoses. Data from the 2009 and 2010 Nationwide Inpatient Sample were obtained for public hospitals and academic medical centers. The final data set included 1,475,457 births in 1373 hospitals in 46 states, with 1,241,255 births to mothers with no prior CDs (primary CD). The prevalence of CD was calculated for all women and all women with prior CDs along with the individual likelihood of CD at each hospital. Lower- and higher-risk women were also identified. Covariates included maternal age, race/ethnicity, and insurance status, and maternal and infant medical conditions. Multilevel logistic regression models were used to quantify the degree to which the variation in CD risk was attributable to hospitals. The average hospital prevalence of CD was 33.0% (95% confidence interval [CI], 32.9%–33.1%) among all births, and the mean prevalence of primary CD was 22.0% (95% CI, 22.0%–22.1%). Average risk of CD and ranges were similar across hospitals of different sizes and location/teaching status for both groups. The individual risk for CD among all women and among those with a prior CD varied by maternal age, race/ethnicity, and insurance status, and medical diagnoses related to pregnancy and delivery. Primary CD prevalence was higher for women with pregnancy-related diabetes (34.5%; 95% CI, 34.1%–34.8%), hypertension (41.4%; 95% CI, 41.1%–41.7%), hemorrhage or placental complications (57.9%; 95% CI, 57.2%–58.6%), fetal distress (58.2%; 95% CI, 57.9%–58.4%), fetopelvic disproportion or labor obstruction (58.4%; 95% CI, 58.0%–58.8%), or maternal age 35 years or older (28.0%; 95% CI, 27.8%–28.2%). In a model without adjustment for maternal diagnoses, the individual risk of primary CD was 0.14 (95% credible interval, 0.12–0.15). The risk of CD varied between 19% and 48% across hospitals. After adjusting for individual diagnoses, sociodemographics, and hospital, the hospital-level variation did not decrease (0.16; 95% credible interval, 0.14–0.18). The likelihood of a parturient having a CD varied between 11% and 36% across hospitals. Cesarean risk among lower-risk women varied across hospitals (0.20; 95% credible interval, 0.18–0.21). The likelihood of a lower-risk woman undergoing CD varied between 8% and 32% across hospitals. Variance did not decrease after adjustment for maternal and hospital factors (0.26; 95% credible interval, 0.23–0.29). Among higher-risk women, hospital-level variance in the likelihood of CD was 0.30 (95% credible interval, 0.28–0.34) before adjustment and 0.25 (95% credible interval, 0.21–0.28) after controlling for maternal diagnoses and hospital factors. The likelihood of a higher-risk woman having a CD varied between 56% and 92% across hospitals. The variation in individual risk of CD across hospitals is not explained by differences in maternal clinical diagnoses. Data on other aspects of individual clinical complexity and hospital factors are needed to enhance understanding of the reasons for variations in the individual likelihood of CD in US hospitals. Because the data did not contain information on parity or gestational age, these factors remain to be examined. More comprehensive data and examination of other factors, including hospital policies, practices, and culture, are needed to more completely determine use of CDs.