This study was designed to assess the impact of selected medical interventions during labor upon cesarean section rates by comparing the maternal and neonatal outcomes of obstetrician- and nurse-midwife-managed low-risk private patients. All patients who delivered at Prentice Women's Hospital in Chicago, Illinois, from January 1, 1987 through December 31, 1990 were evaluated for low-risk criteria to be included in the study. During that time, the nurse-midwives delivered 573 patients and the obstetricians delivered 12,077 patients. Patients with fetal and maternal complications known to increase the cesarean section rate were eliminated from both groups. Eight percent of the nurse-midwife patients and 32% of the physician patients were eliminated, leaving 529 nurse-midwife patients and 8,266 physician patients. These patients were compared for race, parity, age, and birth weight. Information was collected from a perinatal data base and hospital computerized statistics. The rates of cesarean section, administration of oxytocin, analgesia, anesthesia, and infant outcome data were compared by chi-square analysis. Multiple logistic regression analysis was used to assess factors that predicted cesarean section. Nurse-midwife-managed patients had a significantly lower rate of cesarean section (8.5% versus 12.9%; P < .005) and operative vaginal delivery (5.3% versus 17%, P = .0001) than the physician-managed patients. Epidural anesthesia and oxytocin for induction and augmentation were used significantly more frequently in the physician-managed patients. Both interventions were associated with an increased rate of cesarean section. Fetal outcomes in the two groups were not statistically different. Women cared for by nurse-midwives had a lower cesarean section rate, fewer interventions, and equally good maternal and infant outcomes when compared with those cared for by physicians.