Hospitals by day, dispensaries by night: Hourly fluctuations of maternal mortality within Mexican health institutions, 2010–2014

Background Quality of obstetric care may not be constant within clinics and hospitals. Night shifts and weekends experience understaffing and other organizational hurdles in comparison with the weekday morning shifts, and this may influence the risk of maternal deaths. Objective To analyze the hourly variation of maternal mortality within Mexican health institutions. Methods We performed a cross-sectional multivariate analysis of 3,908 maternal deaths and 10,589,444 births that occurred within health facilities in Mexico during the 2010–2014 period, using data from the Health Information Systems of the Mexican Ministry of Health. We fitted negative binomial regression models with covariate adjustment to all data, as well as similar models by basic cause of death and by weekdays/weekends. The outcome was the Maternal Mortality Ratio (MMR), defined as the number of deaths occurred per 100,000 live births. Hour of day was the main predictor; covariates were day of the week, c-section, marginalization, age, education, and number of pregnancies. Results Risk rises during early morning, reaching 52.5 deaths per 100,000 live births at 6:00 (95% UI: 46.3, 62.2). This is almost twice the lowest risk, which occurred at noon (27.1 deaths per 100,000 live births [95% U.I.: 23.0, 32.0]). Risk shows peaks coinciding with shift changes, at 07:00, and 14:00 and was significantly higher on weekends and holidays. Conclusions Evidence suggests strong hourly fluctuations in the risk of maternal death with during early morning hours and around the afternoon shift change. These results may reflect institutional management problems that cause an uneven quality of obstetric care.