Temporal Trends of Maternal Mortality Due to Obstetric Hemorrhage in Chinese Mainland: Evidence from the Population-Based Surveillance Data Between 2000 and 2019

Abstract Objective: To analyze the temporal trends of maternal mortality ratio (MMR) due to obstetric hemorrhage and its specific causes in Chinese mainland from 2000 to 2019, to identify whether the rate of change has accelerated or slowed down during this period, and to find the prior cause of obstetric hemorrhage that needs to be intervened in the future. Methods: Individual information on maternal deaths and total number of live births from 336 surveillance sites across 31 provinces in Chinese mainland was collected from the National Maternal and Child Health Surveillance System between 2000 and 2019. Maternal death was defined according to the World Health Organization's criterion. The final underlying cause of death was confirmed by the national review and was coded according to International Classification of Diseases -10. Linear trends for changes in characteristics of maternal deaths were assessed using linear or logistic models with the year treated as a continuous variable. The MMR and 95% confidence intervals (CI) for regions or causes were estimated by Poisson's distribution. Joinpoint regression was used to assess the accurate temporal patterns. Results: The national MMR due to obstetric hemorrhage was 18.4 per 100,000 live births (95% CI: 15.0–22.2) in 2000. It peaked in 2001 (22.1 per 100,000 live births, 95% CI: 18.3–26.4) and was lowest in 2019 (1.6 per 100,000 live births, 95% CI: 1.0–2.3). For specific regions, the MMR due to obstetric hemorrhage in rural areas and western regions both experienced a slight rise, followed by a rapid decline, and then a slow decline. For specific causes, no change point was found in joinpoint analysis of the national MMR caused by placenta previa, postpartum uterine atony, and retained placenta (the annual percent change was −12.0%, −10.5%, and −21.0%, respectively). The MMR caused by postpartum hemorrhages (PPH) significantly declined by 8.0% (95% CI: 1.9–13.6) per year from 2000 to 2007. The annual percent change of MMR caused by PPH accelerated further to −25.0% between 2007 and 2011, and then decreased to −7.8% between 2011 and 2019. The proportion of maternal deaths due to antepartum hemorrhages increased from 7.6% (8/105) in 2000 to 14.3% (4/28) in 2019. The changes in the proportion of causes were different for maternal deaths due to PPH. The proportion of postpartum uterine atony increased from 39.0% (41/105) in 2000 to 60.7% (17/28) in 2019, and the proportion of uterine rupture also increased from 12.3% (13/105) in 2000 to 14.3% (4/28) in 2019. However, the proportion of retained placenta decreased from 37.1% (39/105) in 2000 to 7.1% (2/28) in 2019. Conclusion: Over the last 20  years, the intervention practice in China has proved that targeted interventions are beneficial in reducing the MMR due to obstetric hemorrhage. However, the MMR has reached a plateau and is likely to increase for some specific causes such as uterine rupture. China needs to develop more effective interventions to reduce maternal deaths due to obstetric hemorrhage, especially for postpartum uterine atony and uterine rupture.