Giving birth at a health-care facility in rural China: is it affordable for the poor?
Qian Long a, Yaoguang Zhang b, Joanna Raven c, Zhuochun Wu d, Lennart Bogg e, Shenglan Tang c & Elina Hemminki f
a. Department of Public Health, University of Helsinki, PO Box 41, Mannerheiminitie 172, FI-00014 Helsinki, Finland.
b. Centre for Health Statistics and Information, Ministry of Health, Beijing, China.
c. Liverpool School of Tropical Medicine, Liverpool, England.
d. School of Public Health, Fudan University, Shanghai, China.
e. Department of Public Health Sciences, Karolinska Institute, Stockholm, Sweden.
f. National Institute for Health and Welfare, Helsinki, Finland.
Correspondence to Qian Long (e-mail: email@example.com).
(Submitted: 06 May 2010 – Revised version received: 22 October 2010 – Accepted: 02 November 2010 – Published online: 02 December 2010.)
Bulletin of the World Health Organization 2011;89:144-152. doi: 10.2471/BLT.10.079434
In China, the number of women giving birth at a health-care facility is used as a target indicator for measuring progress towards improved maternal health.1 The proportion of deliveries attended in health-care facilities varies across geographical areas and according to family wealth: in 2003, about 94% of urban women in China gave birth at a health-care facility. This is 1.4 times the proportion in an average rural area and 3 times that in poor rural areas.2 In addition, giving birth at a health-care facility was four times more common among the richest 20% of women than among the poorest 20% (Z Wu, unpublished data, personal archive, 2010).
A limited ability to pay and high hospital costs have been identified as the major barriers for the rural poor wishing to access health care in China.3 Following the demise of the rural Cooperative Medical Scheme in the 1980s, which occurred with the marketization of the rural economy, large sections of the rural population were left without health insurance cover.4 In addition, China’s health-care system was decentralized in the 1980s and the central budget dropped to 10% of total expenditure.5,6 Health-care facilities now rely on user fees to cover their running costs and the result has been a rapid increase in medical costs.3
Data from the Chinese Ministry of Health show that fee-for-service income accounted for 82% of the total revenue of maternal health-care institutions in rural China in 2002.7 Delivery is the most costly part of maternal care, and expenditure can be especially high for emergency obstetric care.8 Unexpectedly high expenditure on a delivery can push a family into poverty.
In 2003, a new rural health insurance programme, the New Cooperative Medical Scheme, was introduced with the aim of reducing the risk that health-care costs could become catastrophic for some individuals. The scheme operates on a voluntary basis and uses funds pooled from central and local government and from individual contributions.4 The county (typical population: 0.5–1 million) forms the administrative unit of the scheme and the risks associated with illness are shared across the unit. The county government can decide the content, coverage and reimbursement model most appropriate for local conditions, although the Chinese Ministry of Health is responsible for developing overall strategies and policies. In 2008, 92% of the rural population were enrolled in the New Cooperative Medical Scheme.9
The New Cooperative Medical Scheme includes a maternal care benefit package that differs in design and implementation across counties. Usually this package provides reimbursement for delivery at a health-care facility, either as a fixed proportion of expenditure or as a fixed payment. Reimbursement may be the same or different for vaginal and Caesarean delivery.
The aims of this study were to investigate changes in expenditure on facility-based delivery in rural China between 1998 and 2007, to examine the financial burden on households, in particular on poor households, and to identify factors associated with out-of-pocket expenditure on facility-based delivery.
The study was based on cross-sectional data from national household health service surveys conducted in 2003 and 2008 by the Centre for Health Statistics and Information of the Chinese Ministry of Health. For our analysis, only the rural component of the data set was used. Both surveys employed the same four-stage, stratified, random sampling procedure involving counties, townships, villages and households. Ten indicators of socioeconomic development were used to classify each county’s level of development as being in one of four categories: developed, relatively developed, less developed or poor. The probability proportional sampling method was used to randomly select counties for each development category; then, five townships were selected from each county and two villages from each township. Finally, 60 households were selected from each village. In total, 40 212 rural households were surveyed in 2003 and 39 654 were surveyed in 2008.
The health service surveys involved trained township health-care workers carrying out interviews in the selected rural households using structured questionnaires. Each family member answered the questions individually. If one family member was not at home at the time of the survey, another answered the questions on his or her behalf. The questionnaires used in the two surveys had a similar structure and involved similar questions. The 2008 survey included questions about participation in the New Cooperative Medical Scheme and any reimbursement received. The questionnaires were divided into several sections covering: the general demographic and socioeconomic background of the sampled households and family members; the perceived need and demand for health care; and the utilization of and expenditure on health services. The questionnaire included a section on the births that occurred within the 5 years before the survey and these births, which took place between 1998 and 2007, were considered in this study.
The study examined total expenditure on delivery (i.e. reported medical expenditure) within or outside a health-care facility and according to the mode of delivery (i.e. vaginal or Caesarean). In the questionnaires, vaginal delivery included normal vertex delivery and delivery by forceps or vacuum extraction. Facility-based delivery was defined as a birth attended at a health-care facility at a township or higher level. Births outside health-care facilities included delivery at home, in a village clinic or on the way to a health-care facility. Out-of-pocket expenditure on facility-based delivery was calculated as the total expenditure on the delivery minus any reimbursement reported by the women. Out-of-pocket expenditure was used as the main indicator for evaluating the financial burden on households.
The study also investigated the relationship between out-of-pocket expenditure and the following factors: maternal age; maternal educational level (i.e. illiterate, primary school, secondary school, high school or higher); health insurance cover; income category; parity, defined as the number of live births borne by a woman; and the location of the health-care facility (i.e. at a township, county or higher level). In China, education at secondary schools generally continued to the age of 15 years, while education at high schools continued to 17 years. Annual household income in the calendar year that preceded the survey year also included household expenditure on consumables and any savings made during that year. A household’s income category was derived by dividing household income by the number of individuals in the household and using the figure obtained to allocate the household to one of three categories: low-, medium- or high-income. Each category contained one-third of all households.
To help identify changes in the women's demographic and socioeconomic characteristics, the time period 1998–2007 was divided into four parts: 1998–1999, 2000–2002, 2003–2005 and 2006–2007. The χ2 test was used to check for significant differences. The mean values of expenditure on all deliveries and, separately, on vaginal and Caesarean deliveries were compared between the two survey periods using the t-test. Out-of-pocket expenditure on delivery as a percentage of annual household income in 2002 and 2007 was calculated for the three income categories. Expenditure on delivery for all years from 1998 to 2006 was adjusted in line with the annual consumer price index to correct for inflation. Finally, linear regression models were used to study factors associated with out-of-pocket expenditure on delivery at a health-care facility in 2007 by taking the natural logarithm of the observed expenditure. Expenditure was converted into United States dollars (US$) from Chinese yuan using the 2007 exchange rate of 7.4 yuan to the US$. Data were analysed using SAS version 9.1 statistical software (SAS Institute Inc., Cary, United States of America).
The number of women who gave birth in the time periods covered by the two surveys and their demographic and socioeconomic characteristics are listed in Table 1. The age and parity distributions were relatively similar in the four time periods considered. On average, the women's educational level increased over time (P < 0.01). In addition, a substantial increase in the proportion of women with health insurance was observed in the time period 2003–2005. The great majority of those with insurance (i.e. 98%) had enrolled in the New Cooperative Medical Scheme.
Table 1. Demographic and socioeconomic characteristics of women giving birth in rural China, 1998–2007
Location of delivery
The percentage of women who gave birth outside a health-care facility declined rapidly from 45% (535/1181) in 1998 to 10% (148/1481) in 2007, with most of these deliveries occurring at home. In 1998 and 2007, a skilled birth attendant was present at 58% (286/491) and 39% (41/106) of home births, respectively. In 1998, the most common reason for a home birth, as reported by the women, was that they did not feel a need to deliver in a health-care facility (44%; 216/491); the second most common reason was financial difficulties (29%; 144/491). In 2007, the most common reason was a fast delivery (40%; 42/106) and the second most common was financial difficulties (24%; 25/106).
The proportion of women who gave birth in a township hospital remained the same over the study period, but the proportion who delivered at a county or higher-level hospital increased dramatically from 28% (333/1181) in 1998 to 60% (892/1481) in 2007 (Fig. 1). On average, the Caesarean delivery rate increased over time, from 6% (65/1181) in 1998 to 26% (378/1481) in 2007. Although the rate increased in all income groups in all years, women with a high income were always more likely to have a Caesarean delivery than those with a low income (Fig. 1).
Fig. 1. Proportion of women giving birth at or outside of a health-care facility in rural China, by level of facility, and the proportion who had a Caesarean delivery, by household income,a 1998–2007
a Households were divided into three groups of equal size according to the average of their income in 2002 and 2007 combined: low-, medium- and high-income.
Expenditure on delivery
Expenditure on delivery outside a health-care facility increased by 415% over the study period, from US$ 13 in 1998 to US$ 67 in 2007. In each year, facility-based delivery was more expensive than delivery outside a health-care facility and, among facility-based deliveries, a Caesarean delivery was much more expensive than a vaginal delivery (Fig. 2). In 2007, expenditure on a Caesarean delivery was 3.5 times the expenditure on a vaginal delivery. Total expenditure on a facility-based delivery increased by 152% over the study period, from US$ 102 in 1998 to US$ 258 in 2007. Expenditure on a vaginal delivery increased by 226%, from US$ 45 to US$ 146, and on a Caesarean delivery, by 58%, from US$ 326 to US$ 515. For both vaginal and Caesarean deliveries, the greatest increase was seen from 2002 onwards (P < 0.01).
Fig. 2. Expenditurea on delivery at or outside of a health-care facility in rural China, by type of delivery,b 1998–2007
US$, United States dollar.
a Expenditure was adjusted for inflation.
b All deliveries outside a health-care facility were vaginal deliveries.
Since reimbursement data were not available for 2002 and since only around 11% of women had health insurance before 2003 (Table 1), we regarded out-of-pocket expenditure as being equal to total expenditure on delivery in 2002. Between 2002 and 2007, annual household income approximately doubled in all income groups. In these two years, both total and out-of-pocket expenditure on facility-based delivery were higher for medium- and high-income groups than for the low-income group. The increase in total expenditure on facility-based delivery between 2002 and 2007 was around 100% in medium- and high-income groups and 84% in the low-income group. Out-of-pocket expenditure on delivery also increased, but less substantially; the smallest increase was in the low-income group (Table 2). In 2007, out-of-pocket expenditure on delivery consumed 13.1% of annual household income in the low-income group, compared with 9.0% and 5.6% in medium- and high-income groups, respectively. Out-of-pocket expenditure as a percentage of annual household income had declined since 2002 on average, and differences between the income groups had become smaller.
Table 2. Percentage change (PC) and absolute change in percentage (ACP) between 2002 and 2007 in total and out-of-pocket expenditurea on delivery at a health-care facility in rural China, by household income groupb
For vaginal deliveries alone, the greatest increases in total and out-of-pocket expenditure between 2002 and 2007 were observed in the low-income group: 184% and 115%, respectively. However, in both these years, women in medium- and high-income groups spent more than women in the low-income group. Out-of-pocket expenditure on vaginal delivery as a percentage of annual household income did not change greatly in any income group. The highest percentage was 7.4% in the low-income group in 2007.
For Caesarean deliveries alone, the highest total and out-of-pocket expenditures were in the high-income group in both 2002 and 2007. The increase in expenditure between 2002 and 2007 was also greatest in the high-income group: 63% for total expenditure and 40% for out-of-pocket expenditure. In both 2002 and 2007 and in all income groups, out-of-pocket expenditure on Caesarean delivery consumed a high percentage of annual household income; the percentage was particularly high in the low-income group, at 31.8% in 2007. The percentage declined between 2002 and 2007 and differences between the income groups narrowed (Table 2).
Factors affecting out-of-pocket expenditure
Factors affecting out-of-pocket expenditure on facility-based delivery in 2007 were investigated by linear regression analysis using the logarithmic value of expenditure. After adjusting for all variables simultaneously, we found that expenditure on delivery was significantly higher in women aged over 30 years, in those with a high school or higher education, in those with a medium or high household income and in those with a parity of one. In addition, expenditure on a Caesarean delivery was significantly higher than on a vaginal delivery and expenditure on delivery at a county- or higher-level health-care facility was significantly higher than at a township facility. Health insurance cover was associated with significantly lower out-of-pocket expenditure on facility-based delivery. The coefficients derived by the linear regression model are shown in Table 3.
Table 3. Regression coefficients (β) obtained from the linear regression model of the logarithm of out-of-pocket expenditure on delivery at a health-care facility in rural China, 2007
The analysis was repeated for out-of-pocket expenditure on vaginal and Caesarean deliveries separately. Health insurance cover was not associated with lower out-of-pocket expenditure for vaginal delivery, but it was for Caesarean delivery (Table 3). The effects of age, household income, parity and the location of the health-care facility were similar to those found in the analysis of out-of-pocket expenditure on all deliveries.
We found that the proportion of women who gave birth at a health-care facility in rural China increased greatly in the period 1998–2007. Moreover, in 2007 the majority of births occurred at a county- or higher-level health-care facility. There was also a notable increase in the proportion of Caesarean deliveries in rural areas. Total expenditure on a facility-based delivery (both vaginal and Caesarean) increased markedly, particularly between 2002 and 2007, after the New Cooperative Medical Scheme was introduced. Out-of-pocket expenditure also increased, but less substantially. Having health insurance was associated with reduced out-of-pocket expenditure on a facility-based delivery, particularly on a Caesarean delivery. Out-of-pocket expenditure on a facility-based delivery as a percentage of annual household income decreased over the study period. However, the percentage remained high in the low-income group, even in 2007.
The data collected in the two national household health service surveys has previously been shown to be of satisfactory quality in terms of the representativeness of the sample and reliability.10 Nevertheless, there are several limitations. First, interviewees had to recall information over a period of 1 to 5 years. However, most analyses were performed using data for the year immediately before the surveys (i.e. for 2002 and 2007), so that serious recall bias is unlikely. Moreover, any inaccuracies would probably be similar in the two time periods and should not have influenced the trend in expenditure reported in this paper. Second, out-of-pocket expenditure on delivery in 2002 may have been slightly overestimated because, although a small proportion of women had health insurance that year, expenditure was assumed to be equal to total expenditure on delivery owing to the unavailability of reimbursement data. In addition, since the level of reimbursement for a delivery under the New Cooperative Medical Scheme varied across counties and since data on these variations were not available, the study findings should be viewed as preliminary.
Location and type of delivery
The increase in expenditure on facility-based delivery was associated with an increase in the use of higher-level health-care facilities. This suggests that women were highly aware of and accepted the need for a safe delivery. Higher-level health-care facilities, such as those at county and city hospitals, are often regarded by the general population as providing good care, as a result of which they are often overloaded with patients.11 Conversely, township hospitals, particularly in poor areas, are often poorly equipped, have fewer qualified staff and have poor sanitary conditions and hygiene.7 Understandably, women may find them unattractive as a place of delivery. However, overuse of high-level health-care facilities has been found to affect the efficiency of service provision without increasing patient satisfaction or improving health.11 Moreover, the longer distance to a county-level or higher-level hospital12 and the higher cost of receiving health care in this type of facility can place a heavy financial burden on poor families.
The increase in the Caesarean delivery rate also contributed to the rise in expenditure, since a Caesarean delivery was much more expensive than a vaginal delivery. It has been argued that not all Caesarean deliveries are medically necessary.13 Hospitals’ dependence on fee-for-service revenue has fuelled the epidemic of Caesarean delivery in rural China.14 The decentralization of health institution management has given health-care facilities the financial autonomy to generate revenue and manage surpluses. Although the unit price of basic health-care services is still strictly limited, the cost of high-technology investigations and treatments are not controlled effectively and health-care facilities can now make a profit from drug sales to help ensure financial survival.5,6 Consequently, health-care providers have a strong incentive to pursue more expensive and more profitable high-technology procedures.
Expenditure on delivery
We found a marked increase in total expenditure on facility-based delivery and a relatively smaller increase in out-of-pocket expenditure around the period when the New Cooperative Medical Scheme was launched. Health insurance has been advocated as one way of moving closer to universal access to health care, as urged by the 58th World Health Assembly.15 In addition, the Chinese government is committed to increasing funding for the New Cooperative Medical Scheme.16 However, the health-care system has remained heavily dependent on user fees. The coexistence of fee-for-service financing and the government-run voluntary insurance programme may have contributed to the increased cost of delivery and higher expenditure for women. However, causality cannot be concluded from this study. Similar results have been reported in studies of community-based health insurance in other low-income countries, which suggests that such schemes can lead to higher revenues from fee-for-service care.17
In addition, we found that the possession of health insurance, mainly New Cooperative Medical Scheme coverage, was associated with lower out-of-pocket expenditure on Caesarean delivery but not on vaginal delivery. Generally the scheme provides a fixed payment for vaginal delivery. Consequently, while expenditure on vaginal delivery has increased dramatically, the level of reimbursement has remained low, resulting in high costs for women. On the other hand, the scheme views Caesarean delivery as necessitating hospitalization and reimburses accordingly; the reimbursement may cover 40–50% of expenditure. In this way the New Cooperative Medical Scheme has helped reduce out-of-pocket expenditure on Caesarean delivery. However, as long as fee-for-service payment to health care providers continues, the scheme may not mitigate charges for women but may instead contribute to an increase in the income of health-care providers by encouraging Caesarean delivery, even in cases where there is no medical indication.14
Out-of-pocket expenditure on facility-based delivery as a percentage of annual household income decreased during the study period, mainly because rural household incomes increased. However, in 2007, expenditure on facility-based delivery equalled 13% of the mean annual income of low-income households. This is a heavy financial burden on families, as it is above the 10% of annual household income that serves as the threshold for catastrophic health-care spending.18 The burden was even higher for women from low-income households who had a Caesarean delivery, since one-third of the mean annual household income had to be spent on the delivery. Families are at great risk of health-care induced poverty19 and delivery care for the poorest households has become less affordable. In addition, many of the costs of prenatal and postnatal care are not usually covered by health insurance schemes. For example, for families who live some distance from a health-care facility, the costs of transportation, accommodation and food, and the cost in time for the women as well as for their accompanying family members, can be considerable.20
With the aim of enabling convenient access and improving cost-effectiveness,21 we recommend investing in and expanding the capacity of health-care facilities at the township level in rural China. The creators of financial mechanisms for funding health-care facilities and health-care providers should avoid introducing perverse financial incentives. The New Cooperative Medical Scheme should focus on women’s needs and should not encourage health-care providers to promote services that are not medically necessary.
In conclusion, the rise in the proportion of births taking place at higher-level health-care facilities, the increase in Caesarean deliveries and the introduction of the New Cooperative Medical Scheme have all contributed to a rise in expenditure on facility-based delivery. Although recent increases in rural household income have mitigated the financial burden of facility-based delivery for families in general, the burden remains high for the poor.
This study forms part of the CHIMACA project (015396), which is funded by the European Commission INCO Programme and co-ordinated by the National Institute for Health and Welfare, Helsinki. The Centre for Health Statistics and Information of the Chinese Ministry of Health funded data collection and management. Qian Long thanks the China Scholarship Council for supporting her study abroad.
- Maternal and infant health care law of the People’s Republic of China. Beijing: Central People’s Government of the People’s Republic of China; 2005. Available from: http://www.gov.cn/banshi/2005-08/01/content_18943.htm [accessed 17 November 2010]. Chinese.
- Ministry of Health. Report of National Health Services Survey 2003. Beijing: Peking Union Medical College Press; 2004. Chinese.
- Gao J, Qian J, Tang S, Eriksson BO, Blas E. Health equity in transition from planned to market economy in China. Health Policy Plan 2002; 17: 20-9 doi: 10.1093/heapol/17.suppl_1.20 pmid: 12477738.
- Wang Y. Development of the New Rural Cooperative Medical System in China. China World Econ 2007; 15: 66-77 doi: 10.1111/j.1749-124X.2007.00076.x.
- Bogg L, Dong H, Wang K, Cai W, Vinod D. The cost of coverage: rural health insurance in China. Health Policy Plan 1996; 11: 238-52 doi: 10.1093/heapol/11.3.238 pmid: 10160371.
- Yip W, Hsiao WC. The Chinese health system at a crossroads. Health Aff (Millwood) 2008; 27: 460-8 doi: 10.1377/hlthaff.27.2.460 pmid: 18332503.
- Guo Y, Zakus D, Liang H. China: policy and practice of MCH since the early 1990s. Matern Child Health J 2008; 12: 139-48 doi: 10.1007/s10995-007-0305-1 pmid: 18040763.
- Filippi V, Ronsmans C, Campbell OM, Graham WJ, Mills A, Borghi J, et al., et al. Maternal health in poor countries: the broader context and a call for action. Lancet 2006; 368: 1535-41 doi: 10.1016/S0140-6736(06)69384-7 pmid: 17071287.
- Summary of health statistics in 2009. Beijing: Ministry of Health; 2009. Available from: http://www.moh.gov.cn/publicfiles/business/htmlfiles/zwgkzt/ptjty/digest2009/T7/sheet002.htm [accessed 17 November 2010].
- Xu L, Wang Y, Collins CD, Tang S. Urban health insurance reform and coverage in China using data from National Health Services Surveys in 1998 and 2003. BMC Health Serv Res 2007; 7: 37- doi: 10.1186/1472-6963-7-37 pmid: 17335584.
- Eggleston K, Ling L, Qingyue M, Lindelow M, Wagstaff A. Health service delivery in China: a literature review. Health Econ 2008; 17: 149-65 doi: 10.1002/hec.1306 pmid: 17880024.
- Anson O. Utilization of maternal care in rural HeBei Province, the People’s Republic of China: individual and structural characteristics. Health Policy 2004; 70: 197-206 doi: 10.1016/j.healthpol.2004.03.001 pmid: 15364149.
- Klemetti R, Che X, Gao Y, Raven J, Wu Z, Tang S, et al., et al. Cesarean section delivery among primiparous women in rural China: an emerging epidemic. Am J Obstet Gynecol 2010; 202: 65.e1-6-.
- Bogg L, Huang K, Long Q, Shen Y, Hemminki E. Dramatic increase of Cesarean deliveries in the midst of health reforms in rural China. Soc Sci Med 2010; 70: 1544-9 doi: 10.1016/j.socscimed.2010.01.026 pmid: 20219278.
- World Health Assembly resolution 58.33. Sustainable health financing, universal coverage and social health insurance. World Health Organization: Geneva; 2005. Available from: http://apps.who.int/gb/ebwha/pdf_files/WHA58/WHA58_33-en.pdf [accessed 17 November 2010].
- Implementation plan for the recent priorities of the health care system reform (2009-2011). Beijing: Central Committee of the Communist Party of China and the State Council; 2009. Available from: http://www.china.org.cn/government/scio-press-conferences/2009-04/09/content_17575401.htm [accessed 17 November 2010].
- Ekman B. Community-based health insurance in low-income countries: a systematic review of the evidence. Health Policy Plan 2004; 19: 249-70 doi: 10.1093/heapol/czh031 pmid: 15310661.
- Ranson MK. Reduction of catastrophic health care expenditures by a community-based health insurance scheme in Gujarat, India: current experiences and challenges. Bull World Health Organ 2002; 80: 613-21 pmid: 12219151.
- Xu K, Evans DB, Kawabata K, Zeramdini R, Klavus J, Murray CJ. Household catastrophic health expenditure: a multicountry analysis. Lancet 2003; 362: 111-7 doi: 10.1016/S0140-6736(03)13861-5 pmid: 12867110.
- Ensor T, Ronoh J. Effective financing of maternal health services: a review of the literature. Health Policy 2005; 75: 49-58 doi: 10.1016/j.healthpol.2005.02.002 pmid: 16298228.
- Campbell OM, Graham WJ, Lancet Maternal Survival Series steering group. Strategies for reducing maternal mortality: getting on with what works. Lancet 2006; 368: 1284-99 doi: 10.1016/S0140-6736(06)69381-1 pmid: 17027735.