Of the approximately four million women who give birth each year in the United States, nearly 13 percent experience one or more major complications. But the extent to which the rates of major obstetrical complications vary across hospitals in the United States is unknown. We used multivariable logistic regression models to examine the variation in obstetrical complication outcomes across US hospitals among a large, nationally representative sample of more than 750,000 obstetrical deliveries in 2010. We found that 22.55 percent of patients delivering vaginally at low-performing hospitals experienced major complications, compared to 10.42 percent of similar patients delivering vaginally at high-performing hospitals. Hospitals were classified as having low, average, or high performance based on a calculation of the relative risk that a patient would experience a major complication. Patients undergoing a cesarean delivery at low-performing hospitals had nearly five times the rate of major complications that patients undergoing a cesarean delivery at high-performing hospitals had (20.93 percent compared to 4.37 percent). Our finding that the rate of major obstetrical complications varies markedly across US hospitals should prompt clinicians and policy makers to develop comprehensive quality metrics for obstetrical care and focus on improving obstetrical outcomes.