Disparities in health system input between minority and non-minority counties and their effects on maternal mortality in Sichuan province of western China

Maternal mortality ratio (MMR) is one of the most important indicators reflecting the development of a country’s economy, culture, and healthcare system, and is recognized globally. The fifth Millennium Development Goal 5 (MDG5) intended to improve maternal health, and one of the MDG5 targets is to reduce maternal mortality by 75% in 1990–2015 [1]. According to the World Health Organization (WHO) report, approximately 99% of the global maternal deaths occur in developing countries in 2015 [2]. Reducing maternal mortality is a continuing global priority, particularly in developing countries. The transformative new agenda for maternal health has been laid out as part of the Sustainable Development Goals (SDGs): to reduce the global MMR to less than 70 per 100,000 live births by 2030 (SDG 3.1).

The MMR in China has decreased strikingly from 88.8 maternal deaths per 100,000 live births in 1990 to 21.7 in 2014 [3], and China is one of the few countries achieving Millennium Development Goal 5. The main causes of maternal death were obstetric haemorrhage, hypertensive disorders in pregnancy, heart diseases, amniotic fluid embolism, puerperal infection in 1996, and still were obstetric haemorrhage, hypertensive disorders in pregnancy, heart diseases, amniotic fluid embolism, venous thrombosis and pulmonary embolism in 2014 [4]. Such causes are similar between minority and non-minority population nationwide. Despite the fact that the MMR markedly decreased in China, there has been a significant imbalance among different geographic regions (east, central and west regions). In spite of the fact that China government has made efforts to narrow the gaps in maternal mortality, these gaps persist, and the mortality in the western region remains high. One study found that MMR was fundamentally higher in 2010 in the Western area than in other areas: 46.1 per 100,000 live births, compared with 29.1 for the central region and 17.8 for the east [5]. Other studies on three regions of China also found that west region had higher MMR than the central and east region [6, 7]. According to official statistics from China Health and Family Planning Commission Statistical Yearbook 2015, the MMR in the Sichuan of the Western China is around two circumstances higher than the created districts of east region, for example, Beijing, Tianjin and Shanghai.

Western China only takes up 27% of the population of the nation in 2010 census but accounts for 41% of a total number of maternal deaths according to official statistics from China Health and Family Planning Commission Statistical Yearbook 2011. Western China is notable for its ethnic diversity, 71% of China’s ethnic minority population and 46 different minority groups are registered as living there [8, 9]. Sichuan province is a microcosm of the region as a whole, 6.12% of the ethnic minority population, and 14 different minority groups live in this region according to the 2010 census [10]. Most of minority population in Sichuan live together in counties where their ancestors lived for generations, were difficult to be accessed, where the economic and social development are usually lag behind other counties. To facilitate the development of Sichuan province evenly and to provide special help to the minority population who live in those disadvantaged counties, the government has divided counties in Sichuan province into two categories, minority counties and non-minority counties, based on the number of minority residents, number of townships with minority population, historical factors, economic status and geographic location.

Based on the health systems framework for improving maternal, neonatal and child health outcomes developed by United States Agency for International Development (USAID) in 2011 [11], disparity of maternal deaths reflects inequities of the health systems. Between 2002 and 2014, Sichuan government made substantial investments with supporting policies in the health system, some of these have specifically focused on the minority areas in the province. It has critical ramifications to know whether the government endeavours had progressed health system measures and narrowed disparities among minority and non-minority counties, which improvement may impact on reduction or change trend in MMR and explain disparities between minority and non-minority counties, as well as the county level variance. The evidence from China on the relationship between health systems and MMR can be found in some studies, such as higher hospital beds per 1000 population, higher length of highways, higher GDP, higher hospital delivery rate, higher utilization of prenatal care, more village doctors and lower illiteracy rate are persistently associated with lower MMR [12‐16]. However, few study analyzed what we need to know.

While various publications in English had particularly centred around maternal mortality in China, yet just a single study reported the effect of family planning on maternal mortality in Sichuan from 1989 to 1991 [17]. In this study, we used county-level data from Sichuan Province for the years of 2002, 2006, 2010, 2014 to describe the socio-economic environment, maternal health policies and programmes, health human resources, health infrastructure, maternal health care services and their explanation power in reducing ethnic disparities in MMR. This findings may help policy-makers to make well-informed decisions to optimize health system investments,which eventually could effectively reduce regional disparity in MMR.

The study showed that Sichuan government had invested substantial funds in some specific programs to target on the improvement of maternal and child health of the minority areas since 2000. The disparities in MMR and health system between minority and non-minority counties in Sichuan province narrowed over time. But the health human resources, health infrastructure, and maternal health care service in minority counties were still lacked behind of non-minority counties, in addition to a lower social and economic environment than the latter. The study likewise shows that better economic environment, more health human resources and more maternal health care services were essentially connected with reduced MMR. All six health system factors together explained 97.6% ethnic group disparity, with most accountable factors being the hospital delivery and prenatal care. The health system factors together attributed 66.3% county variation, about one-third of county level variation has remained unexplained. The GDP per capita (PGDP) alone explained the 67.9% MMR decreasing.

The fastest declining rate in MMR during 2006 to 2010 could be due to the impact of policies. During this period, health authority in Sichuan province invested a large number of funds in improving hospital delivery, including programme of “Reducing Maternal Mortality and Eliminating Neonatal Tetanus (RMMENT)” from 2000 to 2009, of “Hospital Delivery Subsidy Project in Rural Area (HDSPRA)” started in 2009, of “New Rural Cooperative Medical Scheme (NCMS)” started in 2007. One study found that the RMMENT program has altogether expanded the hospital delivery rate and essentially diminished the MMR in mid-western China [22].

Ethnic minority counties had a lower level in social and economic environment indicators than non-minority counties. The increased per GDP was associated with decreased MMR and explained the most MMR declining trend. Other provinces of China studies also found this association [13, 23]. Economic environment was also significantly associated with other low and middle-income countries, but little related to developed countries [24]. The impact of economic status on MMR may be through other factors such as maternal education, expenditure on health care, quality of care services. Many minority counties are the least developed counties located in mountainous areas. The GDP per capita of minority counties was persistently lower than that of non-minority counties in Sichuan over time.

Increased number of qualified doctors was partly attributable to reduce MMR disparity, has been reported by some previous studies [25‐27], which is consistent with this study. We also found a decreasing trend in qualified doctors among minority counties from 2002 to 2014, in contrast to an increasing trend among non-minority counties. This finding suggested an increasing gap in qualified doctors between minority and non-minority countries, which could cause the ethnic disparity in MMR. Further research in this area is guaranteed.

We found an increased number of health institutes per 1000 population and decreased hospital number among minority counties in Sichuan province and a positive relationship between the institution indicator and MMR, i.e. more health institutions and higher MMR. These findings may suggest some effects of recent government policy in strengthening preventive care, such as “Public health institutions standardisation construction” project. During 2009-2014, the government invested 1.599 billion Chinese Yuan (CNY) for the standardisation construction of 2089 township hospitals, 151 community health centre (station), 13,748 village clinics. Health institutes included township hospitals, community health centre (station) and village clinics. However, most of those health care facilities did not provide curative care nor delivery services. Instead, they collected and managed medical information by request of the health authority. Thus, more maternal deaths may be reported where more health care institutions were in real life. A study by Zanini, RR. et al. also found that increased number of hospitals per 100 thousand inhabitants was associated with a decreased infant mortality rate [28]. No study estimated the association between health institutes and MMR.

The maternal health care services markedly improved over time among all counties, with the faster annual change rate of minority counties than non-minority counties. As a key finding that higher level of hospital delivery rate and antenatal care rate were associated with a lower level MMR, the previous alone clarified the most contrast in MMR between minority and non-minority counties and county variance. Increased hospital delivery rate being associated with decreased MMR has been evidenced by other studies in China [13, 16]. Hospital delivery with skilled attendants may decrease the risk of maternal death. Health authority in Sichuan province invested a large number of funds in improving hospital delivery, including programme of “Reducing Maternal Mortality and Eliminating Neonatal Tetanus” from 2000 to 2009 (RMMENT), of “Hospital Delivery Subsidy Project in Rural Area” started in 2009 (HDSPRA), and of “Improving Hospital Delivery Rate in 31 Minority Counties” started in 2010 (IHDRMC). One study found that the RMMENT program has fundamentally expanded the hospital delivery rate and essentially decreased the MMR in mid-western China [22]. The IHDRMC project has increased the hospital delivery rate in these minority counties from 30.1% in 2010 to 71.0% in 2014 [29]. The New Rural Cooperative Medical Scheme (NCMS) started in 2007 in Sichuan had reimbursement in favour of ethnic minority areas with different rate for caesarean and vaginal deliveries in different level hospitals [30], the NCMS has resulted in increased use of health services and hospital delivery services [31]. But there are still 29% of pregnant women in 31 Minority Counties who refuse to deliver in the hospitals according to our recent unpublished qualitative study. This is not only an economic issue but also issues of traditional culture and beliefs [15]. It is vital to comprehend what cultural and belief factors may impact the utilisation of health services in minority counties so that culturally appropriate health education programs can be run efficiently to promote hospital delivery. We are currently extending our study to identify minority case counties that had the highest and lowest RHD with low or high MMR to develop specific education and treatment program for targeted intervention.

The study has some limitations. First, the maternal death data was collected from the health administrative system, which might have under-reporting bias due to some target setting pressure. So the MMR used in this study could be different from studies where bias adjustment was made. Such bias could be greater in minority counties than non-minority counties in early period because of their much higher MMR, and then reduced more following the improvement of health systems over time as indicated in this study. This suggests that the real gap in the MMR and in the time trends between the two groups could be fractural rather than constant, and the attribution of health system factors to the differences could be variable in percentages accordingly. So the results of this study should be interpreted as descriptive and not as predictive. Further research based on individual data is required on this account. Second, we cannot acquire the individual level data, so the association between the MMR and health system factors cannot be interpreted at the individual level. In spite of these constraints, the study gives vital findings that health system factors and their correlation with reduction of MMR disparity in the past decade.

In summary, this study reviewed maternal health policies and programs on MMR, presented differences between minority and non-minority counties in the MMR and the socio-economic environment, health human resources, health infrastructure, maternal health care services in Sichuan province, and examined explicitly impact of health system indicators on the MMR disparity related to ethnicity. The strong pieces of evidence on ethnic minority related disparity in MMR and on gaps in health system development indicate that efforts to lower MMR in Sichuan province ought to keep on focusing on minority areas later on. Improving hospital delivery and rising economic level of minority areas are still key measures to act on. The finding may reach out to other provinces in western China due to similar geographic, demographic and socio-economic environment [32].

The MMR decreased over the last decade, with the fastest declining rate from 2006 to 2010. The disparity of raw MMR between minority and non-minority counties decreased. The better socio-economic condition, more health human resources and higher maternal health care services rate were associated with lower MMR. Hospital delivery rate mainly determined disparity in MMR between minority and non-minority counties in Sichuan province. The economic mainly determined the change rate of MMR. The results of the present study can help health authorities to understand that the hospital delivery rate is the key healthcare system determinants for the ethnic minority-related disparity in MMR, and rising economic level is key measures to decrease the MMR.