The trends of maternal mortality ratios and cause pattern in 34 Chinese provinces, 1990–2017

Reducing maternal mortality is challenging and has long been a global health priority. It is one of the eight targets in the United Nations (UN) Millennium Development Goals (MDG) framework [1], and a key goal of the Global Strategy for Women’s and Children’s Health launched by the UN Secretary-General in September, 2010 [2]. Over the past two decades, China has seen rapid progress in improving maternal health, pushing down the maternal mortality ratio (MMR; number of maternal deaths per 100 000 livebirths) at an annualised rate of 7.5% per year from 1990 to 2015, one of the fastest decreases in the world [3]. The national MMR fell from 114.2 in 1990 to 85.2 per 100 000 live births in 2000, and to 17.7 in 2015 [3].

The levels and trends of maternal mortality in China has recently been systematically reported using data from the national Annual Report System on Maternal and Child Health (ARMCH) during 1996–2015 [4]. This survey found substantial heterogeneity in MMR at the country level, the annualised rate of decline in maternal deaths across counties of China from 1996 to 2015 ranged from 4.4% to 12.9%, and 2838 of the 2852 (99.5%) counties had achieved the MDG 5 pace of decline. However, it did not document the cause pattern of maternal deaths and its differences in age groups. Understanding the causes to the burden of maternal deaths is a key requirement to further reduce the MMR, and devise targeted intervention policy. More recently, the Global Burden of Diseases, Injuries, and Risk Factors Study 2017 (GBD 2017) provided estimates of maternal causes of death as part of the analysis of all causes of death. In this study, we reported the spatiotemporal trends of maternal mortality by province and age groups, and tracked the key causes contributing to maternal deaths in China during 1990–2017, using data from GBD 2017.

In China, there are more than 17 million livebirths each year [12]. Based on the data sources from GBD 2017, we describe the level, trends and cause patterns of the MMR for each province over a nearly 30-year period. We found substantial variation in MMR across provinces, but most (27/34) showed significant declines greater than the MDG 5 target of 5.5% per year during 1990–2017, and 33 of the 34 provinces have achieved the MMR below the SDG 3.1 target level of 70 in 2017. This systematic analysis uncovers that haemorrhage and hypertensive disorders are the top two specific causes of maternal death currently in China, followed by abortion and miscarriage, indirect causes, and ectopic pregnancy. Mothers older than 40 years have a much higher MMR than those aged 20–29 years.

The drivers of improvement for MMR reduction are variable and multifaceted, including clinical skill, public health and national policy implications. Higher SDI and improved economic conditions usually contribute substantially to decline in MMR [5]. The lowest provincial MMR level in 2017 was 4.2 in Zhejiang, which was similar to those in the most developed countries, including Austria (4.3), Ireland (3.9), Finland (3.6), Italy (3.6), and Denmark (3.6) (https://​maternalhealthat​las.​org/​). However, income per capita in 1996 of China could explain only about 10% of the variation in the annualised rate of decline in MMR over the next two decades; in addition, from 1996 to 2015 the income improvement could explain only about 18% of the changes in MMR [4]. It should be noted that the number of deaths and reductions in MMR need to be considered together to interpret the trends of maternal mortality. For example, meaning and reason for low reduction in MMR for Hong Kong SAR and Tibet are different. It is difficult to achieve a high decline when the number of deaths is already low.

The lower MMR might partly be attributed to the lower fertility rates, but seems more likely to result from interventions introduced by the national Reducing Maternal Mortality and Eliminating Neonatal Tetanus programme of China since 2000. In order to eliminate the inequality of MMR between economically rich and underdeveloped areas, the Chinese government launched this national programme to improve the maternal and neonatal health. It was initially administered to cover the midwestern regions of China, but had become a national policy by 2009. Of this policy, some objectives should be noted, for they might provide effective reference for public health policy makers in other countries (particularly in those with lower economic levels) to devise effective plans for improving maternal health. These objectives consist of: improve obstetric health care at the county, township, and village levels; establish obstetric emergency centres and develop a green channel for obstetric intensive care services in all 2852 counties of China; improve the proportion of in-hospital deliveries and provide financial support for women in rural areas and with low incomes; improve maternal health education; and strengthen supervision of obstetric health care in all medical institutions [4].

In addition, due to the rapid economic development in the past two decades, comprehensive sex education, methods of modern contraception and access to safe abortion might all have contributed to the progress in maternal health and the reduction in maternal mortality. Increased coverage of high-quality antenatal care, and effective identifying and treatment of infectious diseases (e.g., chronic hepatitis B infection), pre-existing chronic conditions (e.g., renal dysfunction, rheumatic, or heart disease) and pregnancy abnormalities (e.g., nutritional deficiencies, hypertensive disorder, and hyperglycemia) are also essential for prevention of maternal deaths [4, 13, 14].

From 2010 to 2015, the percentage of in-hospital delivery increased from 96.3% to 99.7%. During the same period, substantial efforts were also made to offer systematic care for mothers from before to after delivery, and the percentage of pregnancies received these services increased from 80.9% to 91.5% [12]. Despite these impressive improvements, our results showed that five provinces still had MMR significantly higher than the national mean level of 13.6 in 2017, including Tibet (82.7), Xinjiang (41.6), Qinghai (40.5), Guizhou (30.9) and Gansu (23.0). Affected by traditional customs, women in these regions prefer to give birth at home [15]. Besides, low levels of education or even illiteracy have also limited the ability of pregnant women to seek the provision of formal healthcare services [16]. Excluding the educational level, the maternal healthcare medical resources in these areas are also at a disadvantage. Information from the China Health and Family Planning Statistical Year book for 2014 (http://​tongji.​cnki.​net/​kns55/​navi/​YearBook.​aspx?​id=​N2014120147&​floor=​1) revealed that the number of personnel maternal healthcare service per 1,000 persons for the five provinces was lower than other regions in China. The central and local governments should make a concerted effort and devise appropriate interventions to help these provinces. To sum up, cultural practices and low in-hospital delivery rate were reported as major factors behind the high MMR in the western and southwestern areas [13, 14, 17]. Tough natural environments, weaker health services and difficulty in accessing convenient transportation might also contribute to the high MMR in these provinces [4]. Increased coverage of access to and quality of health care could contribute to further declines in maternal mortality in China.

Previous report revealed that more than 95% of maternal deaths in western regions and 76% in eastern regions of China are preventable [13]. Haemorrhage was the leading specific cause of maternal deaths across the provinces in China, accounted for about 25% of all maternal deaths in 2017. It was reported that with advancing in sufficient blood supply, skilful medical procedures and easy transportation, 90% of obstetric haemorrhage caused deaths would be preventable [18]. Continued promotion of policies to control pregnancy-induced hypertension, reduce anaemia and malnutrition, encourage skilled birth attendance and in-hospital delivery, discourage adolescent fertility, and prevent unsafe abortion should lead to sustained maternal health improvement [19‐21]. In addition, special obstetric care focused on fatal conditions of the peripartum and late maternal period, such as pulmonary embolism, cardiomyopathy, and renal complications are also needed to be improved to reduce the preventable maternal deaths. Enhanced data collection with more detailed information on these interventions at the province or even county level will be necessary in the future, which can help the government to create policies with precision.

This study provides comprehensive estimates of the levels and trends of maternal mortality due to different causes, by age and province for the period 1990 to 2017 in China. The limitations of our study should also be pointed out. First, because of sparse data, only nine specific causal categories of maternal death were examined in this report. We were unable to quantify the contributions of other conditions, such as obesity, diabetes, pulmonary embolism, cardiac disorders and hepatitis to the maternal mortality. Second, other potential data sources on maternal health, such as the ARMCH, and the Disease Surveillance Point system administered by the Chinese Centre for Disease Control and Prevention, are key sources of vital statistics for maternal mortality. These data are important for the accurate estimation of levels and trends in maternal mortality of China. However, they did not provide detailed information on cause­specific MMR estimates and the age groups of women. As civil registration systems provide essential information for public health policy devising and disease preventions [22], strengthening of these systems is vital for public health. Third, determination on which deaths of pregnant women should be defined as maternal deaths may influence the diagnosis and estimates of maternal mortality. For example, incidental deaths in which pregnancy had no causal role might be misclassified as maternal deaths. Additionally, in the current stage, China uses all possible medical resources to lower the risk of maternal mortality during pregnancy or within 42 days of termination of pregnancy. Insufficient attention on late maternal mortality may lead to under-estimation of the maternal deaths.

In summary, substantial progress in reducing MMR has been achieved in China during the last two decades, but there is still major work left to do, for disparities in MMR remained striking across provinces. More than 2240 women died during or following pregnancy in 2017, most of which might be preventable deaths. Haemorrhage and hypertensive disorders are the main drivers of maternal death that need intensified focus. Setting associated interventions will need careful consideration of regions that still have MMR significantly higher than the national mean level, such as western China.