In the U.S., childbirth is the most common reason for hospitalization, and cesarean is the most common inpatient surgery 1. In 2011, one in three women gave birth by cesarean delivery—an increase of 60 percent since 19962.
Cesarean deliveries are performed when complications occur during pregnancy or childbirth and a vaginal delivery would threaten the health of the mother or the baby. Cesarean procedures are life saving, but also increase the potential for other risks, such as infection or pain, rehospitalization, problems breastfeeding and future pregnancy complications. Providers, hospitals, insurers, and mothers want to ensure that use of cesarean delivery is not over- or underperformed. In 2014, the American College of Obstetricians and Gynecologists released recommendations to encourage a reduction in non-medically necessary cesarean deliveries3.
Rates of cesarean delivery have been shown to vary significantly across hospitals, which raises concerns about whether the cesarean procedure is being using appropriately according to clinical guidelines. Understanding the drivers of variation in use of the procedure may help improve the quality and value of maternity care in the health care system for the nearly 4 million women and infants who receive childbirth care each year in the United States. Cesarean births cost significantly more—an average of $4,700 without complications—compared to $2,900 for a vaginal birth4. Efforts to reduce unnecessary cesarean procedures, therefore, have the potential to reduce costs to the health care system.
Researchers at the University of Minnesota aimed to explore potential drivers of variation in the use of the cesarean procedure across hospitals, including whether patients’ health conditions or other characteristics could explain the variation in rates across the United States.
This study used national hospital discharge data from the 2009 and 2010 Nationwide Inpatient Sample, which includes all-payer inpatient claims data from a representative 20 percent sample of U.S. hospitals—one of the most comprehensive national sources of information on hospital-based care in the country. In total, the study analyzed data from about 1.5 million births in 1,373 hospitals.
The authors conducted statistical analyses—multilevel logistic regression statistical models—to assess whether variability in hospital rates was explained by differences in patient case-mix across hospitals. The maternal conditions examined were diabetes, hypertension, hemorrhage or placental complications, fetal distress, and fetal disproportion or obstructed labor. Other maternal characteristics examined included age, race/ethnicity, and insurance status (e.g., private insurance, Medicaid, uninsured, etc.) Hospital size and location (rural versus urban), and teaching status were also examined.
Data were analyzed among several groups of women; rates for all women were compared to a group that included only women with no prior cesarean deliveries. Analyzing these two groups separately was important since women who have one cesarean delivery often have cesarean procedures for subsequent deliveries. Two other groups for comparison were also identified—one group of low risk women, which excluded women with preterm delivery (prior to 37 weeks gestation), multiple gestations (e.g., twins), and prior cesarean delivery; and a second high-risk group included those women with preterm delivery, multiple gestation, fetal malpresentation, or prior cesarean section.
Among women with no prior cesarean deliveries, the cesarean delivery rate was 22 percent compared to the entire study population rate of 33 percent.Without taking into account a mother’s health status or medical conditions, the likelihood of an individual having a cesarean delivery varied between 11 percent and 36 percent across hospitals among women with no prior cesarean section.
Among low risk women, the likelihood of undergoing a cesarean delivery varied between 8 percent and 32 percent across hospitals. The likelihood of having a cesarean delivery was significantly higher for high-risk women, which varied between 56 percent and 92 percent across hospitals—a range of 35 percentage points.
For each of these groups, the hospital variance of cesarean delivery rates was similar even after adjusting for maternal characteristics and pregnancy complexity, suggesting that these factors are not driving the variability in rates across hospitals.
However, data about the number of babies a woman has had previously and the gestational age of the baby were not available for analysis and limit the strength of this conclusion. Those two factors can strongly influence the likelihood a woman will deliver via cesarean. In addition, administrative differences or obstetric care policies at the hospital level could not be assessed within this dataset. Discharge data do not contain this level of detail, which constrains the ability to assess the appropriateness of care or possible administrative or clinical explanations for variation across hospitals.
This study was the first analysis of a nationally representative sample, but is consistent with results from data in Arizona and Massachusetts, and among births in U.S. military hospitals. Results from similar analyses in the United Kingdom, however, found approximately one-third of variability in cesarean rates across hospitals was attributable to mothers’ health status and conditions, although these data include information about the number of previous births and gestational age. Findings in the U.S. and United Kingdom may differ due to vast differences in the maternity care systems of the two countries. The United Kingdom’s National Institute for Health and Care Excellence (NICE) encourages healthy women who have had straightforward pregnancies to opt for a home birth or a midwife rather than a doctor. NICE argues that obstetricians are more likely than midwives to over-intervene during a birth, leading to complications that could have been avoided, such as hospital-acquired infections5.
Compared to other high-income countries, the U.S. spends proportionally more on health care, but has worse maternal and infant health outcomes. Understanding the drivers of variation in cesarean rates across hospitals may help inform clinical and policy efforts to reduce unnecessary cesarean sections and support consistent, high-quality maternity care in the U.S.
This study found substantial variation between hospitals in the risk of cesarean among women with no prior cesarean deliveries. Differences in mothers’ health status and health conditions could not explain these differences. These results add urgency to the need for more comprehensive data, and additional research to explore other factors that may be driving variation in cesarean rates, such as hospital policies, practices, and culture.
In addition, variation in cesarean rates highlights the need for greater adherence to guidelines for use of cesarean delivery and other efforts to improve decision-making around childbirth care. Other aspects of care management, including hospital culture, practice patterns, management, training, and organizational change should be taken into consideration when implementing guidelines.
1 Agency for Healthcare Research and Quality. (2014, May). Complication conditions associated with childbirth, by delivery method and payer, 2011. Statistical Brief No. 173.
2 American Congress of Gynecologists and Obstetricians. (2014, February 19). Nation’s Ob-Gyns take aim at preventing cesareans.
3 American Congress of Gynecologists and Obstetricians. (2014, February 19). Nation’s Ob-Gyns take aim at preventing cesareans.
4 National Partnership for Women & Families. (2013, January). The cost of having a baby in the United States.
5 New England Journal of Medicine. (2015, June). A NICE Delivery — The Cross-Atlantic Divide over Treatment Intensity in Childbirth.
For full text of the research article: Kozhimannil, K. B., Arcaya, M. C., & Subramanian, S. V. (2014). Maternal clinical diagnoses and hospital variation in the risk of cesarean delivery: Analyses of a national U.S. hospital discharge database. PLoS Medicine, 11(10), e1001745.
HPM Research Brief is a summary of recent research by faculty and staff in the Division of Health Policy and Management, School of Public Health, University of Minnesota.
HPM research efforts help to influence policy and improve health outcomes across the globe. Visit our website at www.sph.umn.edu/hpm.