655: Variation in cesarean delivery rates among high-risk pregnancies across hospitals in the U.S

Abstract

The rate of cesarean deliveries in the U.S. has peaked in the last decade, now reaching a level over double the upper limit deemed acceptable by the WHO. Studies have also described large unexplained variation in cesarean delivery rates between countries, states, hospitals, and physicians. Our study aims to evaluate the current variation in cesarean delivery rates between hospitals across the U.S., and to evaluate differences in hospital characteristics and high-risk pregnancy factors that could explain the higher cesarean delivery rates seen in certain hospitals. We conducted a retrospective population-based cohort study using the 2011 HCUP-NIS database. We identified all U.S. hospitals with delivery admissions and compared the hospitals with high cesarean delivery rates to those with low/mid rates, in terms of hospital and maternal characteristics, as well as neonatal and maternal outcomes. We used stratified regression to further evaluate the independent impact of each hospital characteristic and pregnancy high-risk factor on the observed disparity in cesarean delivery rates. The cesarean delivery rate in the U.S. rose steadily from 29% in 2003 to 34% in 2012. In 2011, 99% of U.S. hospitals have a cesarean delivery rate beyond 20%. The disparity in cesarean delivery rates between high and low-cesarean-delivery-rate hospitals remained after stratifying by individual hospital characteristics and individual pregnancy risk factor. Specifically, the likelihood of having a cesarean delivery in a woman with a high-risk pregnancy is two-fold (aOR 1.99, CI 1.92-2.06) if she is cared for in a hospital within the high-cesarean-delivery-rate group than in one within the low/mid-cesarean-delivery-rate group. Meanwhile, we found no significant difference in neonatal outcomes or maternal outcomes between patients delivered in a high vs. low-cesarean-delivery-rate hospital. Individual hospitals are in themselves independent risk factors for cesarean delivery, regardless of hospital location, payment source, teaching status and bed size, and regardless of patients’ pregnancy risks and characteristics. Choosing to give birth in a certain hospital can put patients at a two-fold risk of having a cesarean delivery, without benefits in maternal or neonatal outcomes.