Physician cesarean delivery rates and risk-adjusted perinatal outcomes

To compare perinatal outcomes in obstetric practices with high and low cesarean delivery rates.

We conducted a population-based study based on 171295 singleton births in New Jersey in 1996 and 1997. Vital certificate data for each birth were linked to the corresponding hospital discharge records. Nonsubspecialist obstetricians were divided into three groups based on their cesarean delivery rates during the study period: low (less than 18%), medium (18-27%), and high (greater than 27%). Perinatal mortality, rates of birth injury, and uterine rupture were compared among the physician groups after adjustment for differences in patient risks.

Physicians in the frequent cesarean delivery group performed more cesarean deliveries for all major indications. Perinatal mortality rates were comparable among the three physician groups. Low and very low birth weight infants delivered by the high-rate physicians did not have a lower risk of mortality. The risk of intracranial hemorrhage was significantly higher for infants delivered by low-rate physicians than for those delivered by medium-rate physicians (adjusted relative risk [RR] 1.53; 95% confidence interval [CI] 1.07, 2.19). Relative to deliveries by medium-rate physicians, deliveries by low-rate physicians were associated with a lower overall risk of uterine rupture (adjusted RR 0.56; 95% CI 0.34, 0.92). Medium- and high-rate groups had similar occurrences of birth injury and uterine rupture.

Low cesarean delivery rates reduced the rate of uterine rupture and were not associated with increased perinatal mortality. The data suggest a small increase in intracranial hemorrhages in infants delivered by physicians who perform relatively few cesarean deliveries.