Abstract
Objective To compare maternal birth complications early versus late in the academic year and to evaluate the impact of resident work hour limitation on the "July effect." Study Design We conducted a retrospective, population-based cohort study of 628,414 singleton births in Washington State from 1987 to 2012 measuring the adjusted risk of maternal peripartum complications early (July/August) versus late (April/May) in the academic year. To control for seasonal outcome variation unrelated to trainees' involvement in care as well as long-term trends in maternal complications unrelated to variation in trainees' effect on outcomes across the academic year, we employed difference-in-differences methods contrasting outcomes at teaching to nonteaching hospitals for deliveries before and after restriction of resident work hours in July 2003. Results Prior to resident work hour limitation in July 2003, women delivering early in the academic year at teaching hospitals suffered more complications (relative risk [RR] 1.05; 95% confidence interval [CI]: 1.00-1.09; p = 0.03). After July 2003, complication risk did not vary significantly across the academic year except at teaching-intensive hospitals, where July/August deliveries experienced fewer complications (RR: 0.95; 95% CI: 0.92-0.98; p = 0.001). Conclusion Women delivering at teaching hospitals early in the academic year suffered a modest but significant increase in complications before but not after resident work hour reform.