University of Minnesota study finds maternal diagnoses doesn’t explain variation in cesarean rates across US hospitals

You are here

Tuesday, October 21, 2014

In 2011, cesarean delivery was the most common inpatient surgery in the U.S., making up 32.8 percent of all deliveries and more than 1.3 million births. But while cesarean delivery is common, cesarean rates vary 10-fold across hospitals in the U.S. The reasons for the variability are not well understood.


In a new study published today in PLOS Medicine, Katy Kozhimannil, Ph.D., lead author and assistant professor in the School of Public Health at the University of Minnesota and her colleagues at the Harvard School of Public Health, analyzed data from 2009-2010. They looked at more than 1.4 million births in more than 1,300 hospitals, across 46 states, and adjusted rates for maternal diagnoses, socio-demographics and hospital characteristics including size, location and teaching status.


“We found that the variability in hospital cesarean rates was not driven by differences in maternal diagnoses or pregnancy complexity,” said Kozhimannil. “This means there was significantly higher variation in hospital rates than would be expected based on women’s health conditions. On average, the likelihood of cesarean delivery for an individual woman varied between 19 and 48 percent across hospitals.”


Key findings of the study include:


  • The cesarean section prevalence was 33 percent overall, and 22 percent among women with no prior cesareans.
  • Among women without prior cesarean section, an individual woman’s likelihood of cesarean delivery varied between 11 and 36 percent across hospitals.
  • Among lower risk women, likelihood of cesarean delivery varied between 8 and 32 percent across hospitals.
  • Among higher risk women, likelihood of cesarean delivery varied between 56 and 92 percent across hospitals.
  • Hospital variability did not decrease after adjusting for patient diagnoses, socio-demographics, and hospital characteristics.  


Kozhimannil and her colleagues said there are some limitations to the data they used in this study, noting that it does not contain parity (the number of times a woman has given birth before) or gestational age (how far along the pregnancy is), clinical details, notes or reasons for cesarean delivery and hospital-level data on guidelines or policies. They recommend routine data linkages between hospital discharge summaries and other data sources, such as birth certificates.


The authors state that their findings highlight the potential roles of hospital policies, practices and culture in determining cesarean rates.

“Women deserve evidence-based, consistent, high-quality maternity care, regardless of the hospital where they give birth,” Kozhimannil said, “and these results indicate that we have a long way to go toward reaching this goal in the U.S.”