The Effect of "Off Hours" on Maternal and Perinatal Outcomes in a Diverse US Cohort

Objective This study aimed to evaluate the "off-hour effect" on maternal and neonatal adverse events in a large cohort representing U.S. population. Study Design A secondary analysis of the Consortium on Safe Labor (CSL) dataset with 208,695 women and 229,385 deliveries was performed. The study included the deliveries of >= 23 gestational weeks from 19 hospitals in the United States from 2002 to 2008. Babies with congenital anomalies were excluded from neonatal outcomes. We compared maternal and neonatal outcomes of patients delivered during weekdays versus off hours (nights and weekends). The primary outcomes of the study were composite maternal and composite neonatal adverse events. The secondary outcomes were delivery type and individual maternal and neonatal adverse events including maternal death and perinatal mortality rate. Associations between off hours and all the outcomes were analyzed in bivariable and multivariable analyses. The same analyses were performed in strata by indication for admission (spontaneous labor or induction of labor). Results Composite maternal adverse events (6.19 vs. 6.06%, p = 0.41) and maternal death (0.01 vs. 0.01%, p = 0.19) were not significantly different between off hours and weekday groups. In contrast, composite neonatal adverse events (6.91 vs. 5.84%, p < 0.001) and perinatal mortality rate (1.03 vs. 0.77%, p < 0.001) were higher in the off-hour group. After adjusting for confounding variables, only the composite neonatal outcome continued to be associated with off hours (adjusted odds ratio [aOR] = 1.10, 95% confidence interval [CI]: 1.04-1.16). Stratified analyses showed that the off-hour effect for the neonatal composite outcome was not present in those presenting in spontaneous labor (6.1 vs. 5.9%, p = 0.40). Conclusion Off-hour delivery was not associated with severe maternal morbidity and was only modestly associated with severe neonatal morbidity. This association was observed in women undergoing induction, not in those presenting in spontaneous labor. These data draw into question the existence of a clinically meaningful and correctable "off-hour effect" in obstetrics.