Characteristics of Spontaneous Births Attended by Midwives and Physicians in US Hospitals in 2014.

INTRODUCTION: This study compares characteristics and birth outcomes of women attended by certified nurse-midwives/certified midwives (midwives) and physicians in US hospitals in 2014.

METHODS: Data reported in 2014 on the 2003 version of the US birth certificate were examined. Spontaneous vaginal births attributed to midwives and physicians and occurring in hospitals were included. Demographic and risk profiles and adjusted odds ratios for maternal and newborn outcomes were compared by provider type.

RESULTS: Midwives attended 294,604 (12.21%) and physicians attended 2,117,376 (87.79%) of all spontaneous vaginal births occurring in 47 states. Small but statistically significant differences (P < .05) were evident in most characteristics. Substantively, profiles were remarkably similar along all demographic and many medical risk variables. Age, race, and insurance status varied by less than 2 percentage points; education status and Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) participation varied by less than one percentage point. Parity, body mass index, maternal weight gain, tobacco use, trimester care began, and average number of prenatal visits also varied by less than one to 2 percentage points. Midwives attended women with hypertension, diabetes, and infections in proportions similar to physicians. Physicians attended more women with preterm birth and multiple gestation. Midwives attended more women beyond 41 weeks' gestation. There were no differences in 5-minute Apgar scores, neonatal seizures, anomalous neonates, or those no longer living at the time of data collection. Third- or fourth-degree lacerations (odds ratio [OR], 0.81; 95% confidence interval [CI], 0.78-0.84), labor induction (OR, 0.76; 95% CI, 0.76-0.77), and epidural analgesia use (OR, 0.54; 95% CI, 0.53-0.54) were less likely in midwife-attended births. Birth at greater than 42 weeks' gestation was more likely (OR, 2.07; 95% CI, 1.97-2.17) among midwife-attended births.

DISCUSSION: In 2014, midwives practicing in US hospitals attended women with a range of risks similar to those of women attended by physicians. Small but statistically significant differences were noted in most characteristics and should be controlled in comparative studies. This contrasts with earlier reports of midwives caring for disproportionately disadvantaged women compared to physicians.