Utilisation, contents and costs of prenatal care under a rural health insurance (New Co-operative Medical System) in rural China: lessons from implementation

Prenatal care aims to identify and address potential risk factors for adverse pregnancy outcomes in order to reduce pregnancy-related mortality and morbidity [1]. Both the accessibility and quality of care are important to be able to make prenatal care effective. Evidence-based practise guidelines and affordable costs are essential for achieving good quality of prenatal care and avoiding dropout from the care [2, 3].

The new World Health Organization (WHO) prenatal care model recommends a minimum of four visits and provides detailed instructions on the basic components of care for the four-visits for both industrialised and developing countries [4]. Costs of accessing prenatal care including travel costs, service fees and test costs have been identified as a barrier to use care [5]. Hence, including a maternity benefit package into health insurance schemes or providing free care at the point of use has been advocated [6]. In 2001, in developing countries an estimated 68% of women had at least one prenatal visit [7]. Often, the prenatal care fails to meet the standards recommended by WHO due to lack of human resources, equipment and supplies [8].

In China, systematic maternal health care was introduced in the 1980 s. The Chinese Ministry of Health issued a national guideline, the updated version of which recommends starting prenatal visits within the first trimester of pregnancy (< = 12 gestation weeks) and having at least five prenatal visits [9]. It also describes the contents of care for each visit. There is significant disparity in prenatal care use between rural and urban areas. However, the national health services survey data in rural areas from 1993 to 2008 show an increase in early utilisation of prenatal care from 24% to 63% and in the proportion of women having five or more prenatal visits from 11% to 44% [10]. We found only one published study on the content of prenatal care, which shows only 50% of rural women had received components recommended [11].

In rural China, financial constraints have been proposed as the main cause for the under-use of health care in general [12]. Our previous study in western rural areas found that low income women were less likely to have five or more visits than women in the medium and high income groups in 2007 [13]. There is also evidence that the means of payment has an association with prenatal care use. Women participating in a health insurance scheme covering the cost of obstetric care made more prenatal visits than non-insured women [14].

In the mid 1950 s, a collectivism-based rural health insurance system was established, called Co-operative Medical System (CMS) which covered the cost of medical and obstetric care. With the transition from a collective economy to a market system at the end of 1970 s, the CMS collapsed leaving around 90% of the rural population uninsured in the 1990 s [15]. Since 2002, the Chinese government has given a high priority to re-establishing a new rural health insurance system, the New Co-operative Medical System (NCMS), which focuses on reducing risks of catastrophic costs of healthcare. The NCMS is organized, guided and supported by central and local governments and involves voluntary participation with a flat rate premium [16]. It operates at the county level and schemes vary across the counties. In 2007, the overall participation rate of total rural residents was 86% [17]. To our knowledge, in most counties the NCMS includes a maternity benefit package providing either a proportional or fixed amount of reimbursement for facility-based delivery, both for vaginal and caesarean delivery. Only a few counties have included prenatal care into the NCMS maternity benefit package.

This study aims to describe prenatal care use, content and costs of care in one county where prenatal care is included in the NCMS and two counties where it is not. It also explores the perceptions of stakeholders on the prenatal care benefit package in order to understand the strengths and weaknesses of the approach in the context of rural China and to draw lessons from early implementation.

Both quantitative and qualitative methods were used in this study.

Whether in the county that included prenatal care in the NCMS, or counties that did not, a substantial percentage of rural women made early and five or more prenatal visits. However, in the study counties, some basic components of care were not provided according to the national guideline; instead, ultrasound test was performed the most frequently. A low proportion of the NCMS members claimed reimbursement: the main reason for not claiming was due to the care being sought for unauthorised pregnancies. In the low income group, the out-of-pocket expenditure for prenatal care accounted for a very high proportion of women's annual income, especially in the relatively poor county.

In this study, there were several limitations. Firstly, data on the content of prenatal care received by women and costs of care are subject to recall bias, though a one-year recall is unlikely to cause serious bias. Secondly, the non-response rates are around 40% in study counties. One of the major missing reasons was women being absent due to rural-to-urban migration. Those migrant women might make some or all of their prenatal visits at rural facilities, but they are likely to differ from rural resident women. We also did not have information about other non-respondents to assess if they differed from respondents on any important variables. In addition, the participants in focus group discussions were recruited by township doctors. Those women might have more contacts with doctors and thus had better awareness of prenatal care use. Generally, the study design and descriptive statistical analyses in this paper cannot be used to draw conclusions on the causality between availability of the prenatal care package in the NCMS and the utilisation of care. The results should therefore be viewed as preliminary and generalisations should be made with caution.

In the three counties, the proportions of women seeking early and adequate prenatal care were higher than national average level in rural China in 2008 [10] and reported levels in other developing countries [7, 21], although the definition of early and adequate prenatal care use varied from those studies. This suggests that in general women's awareness and acceptance of the need for prenatal care use was relatively high. This may be associated with the high level of education in study counties [13]. Compared to women having only one child, women having two or more children were less likely to use prenatal care, which is consistent with other studies in China and worldwide [5, 22].

Prenatal care as part of an insurance benefit package has been successful in increasing the number of visits in the Philippines and some African countries [23, 24], although in the latter study its effectiveness depended on the context in terms of general prenatal care costs and utilisation rates. In our study, part of prenatal care included in the NCMS also aimed to improve prenatal care use, but its implementation was affected by several factors. The first is a factor specific to China: women having unauthorised pregnancies could not benefit from this package because of the national family planning policy (known as the 'one-child policy'). The qualitative study found that the prenatal care included in the NCMS did not influence the decision to make prenatal visits among women having only one child since the demand was high in this group. By contrast, women having unauthorised pregnancies used less prenatal care. Previous studies in rural China have reported a family financial strain created by the substantial fine for unauthorised births that discouraged couples from purchasing both prenatal and obstetric care [25, 26]. A financial benefit may encourage this marginal group to use prenatal care at the recommended time and tests. But there was divergence with the NCMS policy, creating a dilemma for maternal healthcare policy makers.

A second possible reason for the failure of the prenatal care benefit package to improve prenatal care use was that it was not well understood by the women. In our study, less than half of the NCMS members knew that prenatal care was included in the NCMS. Those women who knew that the reimbursement was available for prenatal care did not know the details of the policy and they showed distrust of doctors with regard to the amount of reimbursement that would be provided. A review on general health services delivery in China reported that many patients complained about unclear information about services they received and a few of them or their relatives had open conflict with health providers [27]. Studies in both industrialised and developing countries have found that good patient-provider interaction was associated with increased women's satisfaction [28, 29], which was considered as one of the major factors influencing the use of prenatal care and the effect of care [30]. The lack of clear and effective communication suggested in this study may not only undermine the provider-user relationship, but also potentially affect health seeking behaviour if patients are unwilling to seek care due to lack of trust. Lack of knowledge about benefits available and claim processes have also been identified as barriers both to uptake of general health services covered by health insurance and to successfully claiming against insurance by women in India [31] as well as other studies in China [32‐34].

Insurance-related increase of unnecessary service provision has drawn attention from policy makers internationally. In Taiwan, the use of more expensive prenatal care services notably increased after the implementation of national health insurance [35]. Our study found that the NCMS provided reimbursement for the specific tests and number of tests recommended by the national guideline. This is a cost-containing measure as well as a quality control measure. However, the survey in the three counties found that the basic components of care, such as haemoglobin tests and urine tests, were not carried out following the guideline, and this was particularly so in the county having the prenatal care coverage in the NCMS; while 90% of women received ultrasound tests two or more times. We did not investigate the detailed knowledge of the guideline among health care providers. It may be that ultrasound test is seen as the most important channel to identify risk during pregnancy. However, our qualitative study indicated that some doctors recommended more expensive tests to women, which suggests that considerations other than clinical needs are influencing their provision.

Since health sector reforms were launched in mid-1980, the government allocation to health care has shrunk and the health care system has become heavily dependent on fee-for-services financing. A fee-for-services payment method acts as an incentive to health care providers to over-provide services [36]. Public health care has suffered from a lack of funding and effective guidance from the central and local government in China. Healthcare providers have little incentive to provide non-profit services. Instead, they recommend more or more expensive services than clinically necessary. A study in other areas of China reported that the frequent use of ultrasound tests was to generate more revenue for salaries and pensions of hospital staff [37]. The failure to follow the guidelines for providing prenatal care raises questions about the quality of care. Meanwhile, providing expensive and possibly unnecessary services also increases women's financial burden due to prenatal visits.

According to our results, in RC County, the out-of-pocket expenditure for prenatal care was lower than other two counties. However, given that only a small proportion of women claimed the reimbursement and the amount was small, the NCMS may make little contribution to the lower expenditure for prenatal care. The lower expenditure in this county may be attributed partly to the fact that free prenatal visits are offered for all pregnancies funded by the county government and partly to the fact that fewer tests including some basic tests and ultrasound test were performed than other two counties. In the three counties, the out-of-pocket expenditures as a proportion of women's annual income in the low income group were high. When indirect costs are added including transportation for prenatal visits, the expenditures are substantial for the rural poor. Including prenatal care in the NCMS offers an opportunity to reduce the financial burden caused by prenatal visits, especially for the poor, but our study suggests that greater attention to the relationship between financial protection, quality of provision and provider payment mechanisms is necessary to make this opportunity a reality.

The lessons from our study have important policy implications. Since 1994, the orientation of the Chinese family planning has shifted from a focus on birth control to an integration of birth planning with quality and safety of reproductive health care, poverty alleviation and economic development [38]. The NCMS, a rural health insurance scheme, is intended to contribute towards addressing specific health needs of the target population. Hence, it has to consider how to address current conflicts between national family planning policy and NCMS policy in order to protect health of both authorised and unauthorised pregnancies. Moreover, where inclusion of pre-natal care in the NCMS is adopted, dissemination of the benefit package of the NCMS scheme and reimbursement procedure should be strengthened. In addition, either lack of related knowledge or revenue driven behaviour contribute to questionable quality of prenatal care in rural areas. Further research on factors influencing provider behaviour is needed in order to generate evidence for intervention development. Although one factor affecting the implementation of rural health insurance as a mechanism for improving utilisation and quality of provision was unique to China (the one-child policy), the study also offers lessons for international policy and research. In particular the study illustrates the importance of taking a health systems approach to evaluation that links utilisation and provision of services with the wider context of policy, patient information and demand, and provider motivations.

In rural China, prenatal care use was high, but the contents of care provided did not follow the national guideline and more expensive tests were recommended by doctors. A low proportion of NCMS members claimed the reimbursement available, with the most frequent reason for not claiming being the unauthorised status of the pregnancy for which care was sought. The cost of pre-natal care was substantial for rural poor whether or not it was included in the NCMS benefit package. Further efforts to increase access to affordable, quality prenatal healthcare are therefore needed.

The authors declare that they have no competing interests.