Utilisation of maternal health care in western rural China under

a new rural health insurance system (New Co-operative Medical

System)
Qian Long
1
, Tuohong Zhang
2
, Ling Xu
3
, Shenglan Tang
4
and Elina Hemminki
5
1 Department of Public Health, University of Helsinki, Helsinki, Finland
2 School of Public Health, Peking University, Beijing, China
3 Centre for Health Statistics and Information, Ministry of Health, Beijing, China
4 Liverpool School of Tropical Medicine, Liverpool, UK
5 National Institute for Health and Welfare, Helsinki, Finland
Summary objective To investigate factors inuencing maternal health care utilisation in western rural China
and its relation to income before (2002) and after (2007) introducing a new rural health insurance
system (NCMS).
methods Data from cross-sectional household-based health surveys carried out in ten western rural
provinces of China in 2003 and 2008 were used in the study. The study population comprised women
giving birth in 2002 or 2007, with 917 and 809 births, respectively. Correlations between outcomes and
explanatory variables were studied by logistic regression models and a log-linear model.
results Between 2002 and 2007, having no any pre-natal visit decreased from 25% to 12% (differ-
ence 13%, 95% CI 1017%); facility-based delivery increased from 45% to 80% (difference 35%, 95%
CI 2937%); and differences in using pre-natal and delivery care between the income groups narrowed.
In a logistic regression analysis, women with lower education, from minority groups, or high parity
were less likely to use pre-natal and delivery care in 2007. The expenditure for facility-based delivery
increased over the period, but the out-of-pocket expenditure for delivery as a percentage of the annual
household income decreased. In 2007, it was 14% in the low-income group. NCMS participation
was found positively correlated with lower out-of-pocket expenditure for facility-based delivery
(coefcient )1.14 P < 0.05) in 2007.
conclusions Facility-based delivery greatly increased between 2002 and 2007, coinciding with the
introduction of the NCMS. The rural poor were still facing substantial payment for facility-based
delivery, although NCMS participation reduced the out-of-pocket expenditure on average.
keywords maternal health care, New Co-operative Medical System, China, cross-sectional household-
based health survey
Introduction
China is on track to achieve the Millennium Development
Goal of a reduction in both maternal and child mortality.
The maternal mortality ratio (MMR) has declined from 89
per 100 000 live births in 1990 to 47 per 100 000 live
births in 2005. However, the achievements show signi-
cant regional disparity. In poor western rural areas, the
MMR was four times that of urban areas and twice that of
average rural areas (Jing & Kaufman 2008).
Studies in many developing countries show a positive
association between reductions in maternal mortality and
utilisation of maternal health care (Adam et al. 2005;
Ekman et al. 2008). Utilisation is dened as attendance at
pre-natal and post-natal care and delivery at a health
facility. In China, inadequate use of maternal health care,
especially delivery care, has usually been considered a
major reason for persistently high rural maternal mortality
(Anson 2004; Short & Zhang 2004). In 2003, in the
poorest rural areas, only 16% of women gave birth at
health facilities compared to 93% of women in urban areas
(Ministry of Health 2004).
Financial difculty and a lack of perceived benets from
seeking care are seen as the main causes for the underuse of
Tropical Medicine and International Health doi:10.1111/j.1365-3156.2010.02602.x
volume 15 no 10 pp 12101217 october 2010
1210 2010 Blackwell Publishing Ltd
maternal health care in China (Bogg et al. 2002; Kaufman
& Jing 2002; Wu et al. 2008). In 2003, the most important
reason given for home delivery was that a facility-based
delivery was too expensive (Ministry of Health 2004). The
willingness to meet this cost is further undermined by the
perceived lack of benet from seeking care, with percep-
tions inuenced by age, education, history of pregnancy-
related activities, cultural beliefs and family support
(Li 2004; Short & Zhang 2004; Wu et al. 2008).
In the 1970s, China had a nationwide rural health
insurance system characterised by its collective nancing,
called the Co-operative Medical System (CMS). It covered
90% of the rural population, ensuring universal access to
essential medical care. However, this system collapsed in
the 1980s with the economic transition (Feng et al. 1995),
meaning health care became less affordable in rural China
(Liu et al. 2008). In the early 1990s, the Chinese govern-
ment tried to re-establish CMS, but the scheme could not
be sustained in poor areas because of insufcient nancial
support (Carrin et al. 1999).
Since 2003, a new health insurance system for rural
people, the New Co-operative Medical System (NCMS),
has been implemented that aims to provide nancial risk
protection for the rural population in regard to cata-
strophic disease. It is nanced by individual contributions
and by central and local government. The NCMS operates
with a voluntary enrolment; it is administered at county
level, although rules and policy guidelines are set by central
government (Wang 2007; You & Kobayashi 2008). In the
initial stage of the NCMS, rural residents had little
willingness to participate in the schemes. Participation was
associated with relatively low household income, higher
education, number of sick family members, potential
benets from premium and co-payment costs (Zhang et al.
2006; Wang et al. 2008). The participation rate has
consistently increased, however, reaching 86% in 2007
(You & Kobayashi 2008).
The NCMS has a maternal health care benet package.
Its design and implementation vary across counties. The
package usually provides reimbursement either as a xed
proportion of expenditures or a xed amount for facility-
based delivery, both for normal delivery and caesarean
section. Women pay for pre-natal care, including physical
check up by a doctor, laboratory tests, drugs and
treatments. To our knowledge, only a few counties have
included part of pre-natal care into the NCMS maternal
health care benet package.
This study focuses on the demand side of maternal
health care, aiming to investigate factors inuencing the
use of maternal health care in western rural China and its
relation to income before (2002) and after (2007) intro-
ducing the NCMS.
Data and methods
Data
We analysed regionally representative data from ten
western provinces gathered in a health survey carried out in
2003 and 2008. The survey was developed by the Centre
for Health Statistics and Information (CHSI), which
operates under the Chinese Ministry of Health. In both
surveys, multistage sampling was used in each of the ten
provinces to randomly select thirty rural townships, two
villages in each township and thirty households from each
village, using residence register data (HuKou), yielding
18 000 households. The survey was carried out by trained
township health workers using a structured questionnaire.
Where households refused to be interviewed or no-one was
home at the time of survey (10%), alternative households
were selected. The questionnaire included questions on the
general demographic and socioeconomic background of
the sampled households and family members, the perceived
need and demand for health care, and the utilisation and
expenditures of health services. One section on pregnancy-
related activities was exclusively for women of reproduc-
tive age (1549). Only data for rural women who gave
birth in 2002 or 2007 are reported here.
Methods
Four indicators of the use of maternal health care as
recommended in the Maternal and Infant Health Care
Law of the Peoples Republic of China were examined:
(i) use of pre-natal care, that is, having at least one
pre-natal visit; (ii) initiation of pre-natal care within the
rst trimester of pregnancy (12 gestation weeks); (iii) ve
or more pre-natal visits; and (iv) delivery at a health
facility, dened as giving birth at a township or higher level
health facility. In addition, out-of-pocket expenditure for
facility-based delivery was used as an indicator to evaluate
the nancial consequence for rural households.
The explanatory variables were determinants of mater-
nal health care utilisation identied in previous studies:
age, womens education attainment (illiterate, primary
school, secondary or higher), ethnicity (Han or minorities),
parity (number of live births), history of abortions
(miscarriage or induced abortion) or stillbirths, distance to
the nearest health facility (e.g. village clinic, township or
higher level health facility), income level and NCMS
participation. Income level was estimated using annual per
capita income. The reported gross household income was
divided by the number of individuals in the households and
grouped into three income categories (low, medium and
high), each containing a third of the households
interviewed.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
2010 Blackwell Publishing Ltd 1211
Centre for Health Statistics and Information (CHSI)
allowed data analysis (made by QL) only in CHSI
premises to protect data condentiality. Cross-tabulation
was used to compare utilisation of pre-natal and delivery
care between 2002 and 2007. The 95% condence
intervals for the differences in maternal health care use
between the time periods were tested. The range method
and chi-square trend test were used to examine the
variation in trends for maternal health care use by
increasing income. Logistic regression models were used
to analyse associations between the use of pre-natal and
delivery care and explanatory variables separately in
2002 and 2007. Furthermore, we did the analysis
combining the datasets of 2002 and 2007 adding time
(year) as a dummy variable. Based on womens self-
report data, we calculated the mean delivery expenditure
at the health facility, the mean out-of-pocket expendi-
ture, and the mean annual household income among the
different income groups in 2002 and 2007. In addition,
in 2007, the out-of-pocket expenditure of facility-based
delivery was transformed into natural logarithms of the
observation value to address the positive skew of the
expenditure data. Finally, a linear regression model was
used to study the correlation between the logarithmic
value of the out-of-pocket expenditure of delivery care
and the NCMS, adjusting for covariates. If the estimated
coefcient of the variables was negative, it indicates a
correlation with reduced out-of-pocket expenditure; a
positive coefcient indicates a correlation with increased
expenditure.
Results
The number of respondents included in the study who gave
birth in the calendar year prior to the survey was 917 in
2002 and 809 in 2007. Table 1 shows that the two groups
were relatively similar in regard to age, ethnicity and
parity. The most signicant change between 2002 and
2007 was the increase in the proportion of women
participating in the NCMS: from only a small percentage
to almost everyone. Illiteracy among women decreased
from 2002 to 2007, and over half of the women in 2007
had received at least a secondary education or more. The
numbers of those reporting a history of abortion or
stillbirth were almost twice as high in 2007 than in 2002.
Utilisation of maternal health care
Utilisation of pre-natal and delivery care increased from
2002 to 2007 (Table 2). The proportion of women not
using any pre-natal care decreased, and the proportion
making many pre-natal visits increased, with more women
starting care early. Home deliveries declined from over half
in 2002 to a fth in 2007, with 46% of women choosing to
deliver at either county or higher level health facilities.
Most women had a normal vaginal delivery without the
use of instruments, but there was a notable increase in the
use of caesarean section (from 6% to 17%).
In both study years, women with higher income used
more pre-natal care and gave birth more often in health
facility than women with lower income (Table 3). The
ratio (RR) and difference (RD) between the high- and low-
income groups for use of pre-natal and delivery care was
smaller in 2007 than in 2002. However, with the exception
of the start time for pre-natal care, the differences between
the income groups were statistically signicant even in
2007.
Table 1 Demographic and socioeconomic characteristics of wo-
men giving birth, by year % (N)
2002
(N = 917)
2007
(N = 809) P value
Age
1524 37.8 (347) 38.9 (315) 0.06
2529 35.7 (327) 30.7 (248)
3049 26.5 (243) 30.4 (246)
Education*
Illiterate 22.6 (206) 14.6 (117) <0.01
Primary school 37.1 (339) 31.7 (255)
Secondary school
or higher
40.3 (368) 53.7 (432)
Ethnicity
Han 55.6 (510) 52.9 (424) 0.27
Minority 44.4 (407) 47.1 (377)
Parity
1 46.1 (422) 50.4 (405) <0.05
2 38.8 (355) 39.0 (313)
3+ 15.2 (139) 10.6 (85)
History of abortion
or stillbirth
11.7 (105) 19.7 (158) <0.01
Distance to the nearest health facility
2 km 46.9 (430) 42.7 (345) 0.12
34 km 32.8 (301) 33.5 (271)
5 km 20.3 (186) 23.9 (193)
NCMS participation 5.4 (49) 92.8 (751) <0.01
P value refers to the difference in the distribution between the
years.
*Four women in 2002 and ve women in 2007 with missing
values.
Eight women in 2007 with missing values.
Parity: number of living children; one woman in 2002 and six
women in 2007 with missing values.
Twenty-three women in 2002 and six women in 2007 with
missing values.
Two women in 2002 with missing values.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
1212 2010 Blackwell Publishing Ltd
Factors related to the utilisation of maternal health care
After adjusting for other factors, in 2007 age was positively
related to any pre-natal visit and facility-based delivery,
but the positive association with ve or more pre-natal
visits was not statistically signicant (Table 4). Education
was positively associated with the use of pre-natal and
delivery care. Minority women were less likely to have
early or adequate pre-natal visits, but there was no
difference in delivery care utilisation. Having existing
children was negatively associated with the pre-natal and
delivery care utilisation. Women who reported a history of
abortion or stillbirth were less likely than those without to
start pre-natal visits early. High income was related to the
use of any pre-natal care and many visits, but not to early
start. Women with medium and high income more often
had facility-based delivery than low-income women, but
the difference was not statistically signicant. NCMS
participation was positively related to using any pre-natal
care. It was also related to making early or adequate
pre-natal visits or having facility-based delivery, but the
odds ratios were not statistically signicant. Finally, the
number of pre-natal visits was positively associated with
giving birth at a health facility.
When we combined the datasets of 2002 and 2007
adding time (year) as a dummy variable, we found very
similar results to those for 2007 alone. However, shorter
distance to health facility was related to ve or more
pre-natal visits (data not shown).
Table 2 The utilisation of maternal health
care, by year, % (N)
2002
(N = 917)
2007
(N = 809) RD (95% CI)
Number of pre-natal visit(s)
0 25.3 (232) 11.6 (94) )13.7 ()10.0, )17.3)
14 47.8 (438) 52.3 (423) +4.5 ()0.3, +9.3)
5+ 26.9 (247) 36.1 (292) +9.2 (+4.7, +13.6)
Start of pre-natal visit*
No visit 25.4 (232) 11.7 (94) )13.7 ()10.0, )17.3)
12 gestation weeks 35.2 (321) 55.0 (440) +19.8 (+14.6, +24.0)
>12 gestation weeks 39.4 (361) 33.3 (267) )6.1 ()1.7, )10.9)
Delivery place
Home
Without skilled attendance 31.0 (282) 13.0 (103) )18.0 ()14.1, )21.8)
With skilled attendance 21.5 (196) 6.3 (50) )15.2 ()12.0, )18.4)
Township health facility 19.7 (179) 34.0 (270) +14.3 (+9.6, +18.1)
County or higher level 25.4 (231) 45.8 (363) +20.4 (+15.1, +24.1)
Others (e.g. village clinics or on
the way to health facility)
2.4 (22) 0.9 (7) )1.5 ()0.2, )2.9)
Delivery mode
Normal delivery 91.3 (834) 80.0 (635) )11.3 ()9.0, )15.6)
Forceps vacuum 2.5 (23) 2.8 (22) +0.3 ()1.4, +1.9)
Caesarean section 6.2 (57) 17.2 (137) +13.0 (+7.6, +13.9)
RD, the difference of the use of maternal health care between the years.
*Three women in 2002 and eight women in 2007 with missing values.
Seven women in 2002 and sixteen women in 2007 with missing values.
Three cases in 2002 and fteen cases in 2007 with missing values.
Table 3 The utilisation of maternal health care, by income group,
by year
Income group (%)
RD (%) RR P value Low Medium High
Use of any pre-natal care
2002 64.5 75.7 83.8 19.3 1.3 <0.01
2007 82.4 90.2 93.3 10.9 1.1 <0.01
Start of pre-natal visit 12 gestation weeks
2002 43.3 44.5 52.8 9.5 1.2 <0.05
2007 59.7 61.5 65.2 5.5 1.1 0.22
5+ pre-natal visits
2002 16.1 26.1 38.3 22.2 2.4 <0.01
2007 27.7 35.6 46.2 18.4 1.7 <0.01
Delivery at health facility
2002 28.6 44.9 61.1 32.5 2.1 <0.01
2007 70.2 81.8 87.7 17.5 1.3 <0.01
RD, the difference of the use of maternal health care between the
high-income group and the low-income group; RR, the ratio of the
use of maternal health care between the high-income group and
the low-income group; P value, chi-square test for trend to test sta-
tistical difference of the use of maternal healthcare by income group.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
2010 Blackwell Publishing Ltd 1213
Facility-based delivery cost and the NCMS
In 2002, out-of-pocket expenditure for facility-based
delivery was not asked. Given that most rural people did
not have any health insurance before 2003, we assumed
that total delivery expenditure was equal to the out-of-
pocket expenditure. From 2002 to 2007, total expenditure
of facility-based delivery increased by 48% and 53%
among the low- and medium-income groups, 76% among
the high-income group (Table 5). There was almost no
increase in out-of-pocket expenditure among
the low-income group, a modest increase among the
medium-income group, and a bigger increase among the
high-income group. In 2007, in all income groups, the
annual household income almost doubled compared to
2002. The average out-of-pocket expenditure as a per-
centage of the annual household income decreased from
2002 to 2007. In both years, it was higher among the low-
income group than among the medium- and high-income
groups, even though the difference decreased.
Table 4 Determinants of the utilisation of pre-natal and delivery care, odds ratios (95% condence intervals) from multivariable logistic
regression analysis, adjusting for other variables, 2007
Use of any pre-natal
care
Start of pre-natal
visit 12 gestation weeks 5+ pre-natal visits
Delivery at modern
health facility
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
Age (1524)
2529 2.50 (1.314.74) 0.88 (0.591.32) 1.40 (0.942.07) 1.10 (0.661.86)
3049 2.28 (1.134.62) 0.77 (0.481.24) 1.45 (0.902.33) 2.63 (1.394.97)
Education (illiterate)
Primary school 1.81 (0.993.29) 1.31 (0.782.21) 2.02 (1.103.72) 1.86 (1.093.20)
Secondary school or higher 3.49 (1.826.67) 1.72 (1.042.86) 4.23 (2.367.61) 3.66 (2.066.50)
Ethnicity (minorities)
Han 1.58 (0.952.63) 1.44 (1.032.03) 1.45 (1.042.02) 0.99 (0.641.53)
Parity (3+)
1 3.26 (1.467.27) 2.14 (1.124.08) 1.92 (0.983.77) 3.29 (1.547.05)
2 2.25 (1.134.47) 1.86 (1.043.33) 1.63 (0.883.04) 1.29 (0.722.69)
History of abortion stillbirth (No) 0.69 (0.381.24) 0.64 (0.420.97) 1.23 (0.821.83) 1.26 (0.732.19)
Distance to health facility (5 km)
2 km 0.74 (0.401.37) 0.95 (0.621.45) 1.05 (0.701.58) 0.76 (0.451.31)
34 km 0.68 (0.361.28) 0.87 (0.561.35) 0.80 (0.521.24) 0.74 (0.421.29)
Income (low)
Medium 1.83 (1.063.17) 1.03 (0.701.53) 1.29 (0.881.89) 1.42 (0.892.26)
High 2.01 (1.083.99) 1.05 (0.701.59) 1.49 (1.012.20) 1.49 (0.872.55)
NCMS participation (No) 2.94 (1.356.25) 1.67 (0.883.23) 1.67 (0.873.12) 1.45 (0.693.03)
Pre-natal visit (5+)
0 0.06 (0.030.13)
14 0.34 (0.190.58)
The reference group is shown in parentheses.
Table 5 The expenditure of facility-based delivery, by income group, by year
Low Medium High
2002 2007 2002 2007 2002 2007
Delivery expenditure at health facility, mean (1) 729 1081 889 1366 986 1739
Out-of-pocket delivery expenditure, mean (2)* 729 766 889 1037 986 1456
Annual household income, mean (3) 2655 5407 4902 11884 14770 29723
Delivery expenditure at health facility, % of income (4) = (1) (3) 27.5 20.0 18.1 11.5 6.7 5.8
Out-of-pocket delivery expenditure, % of income (5) = (2) (3) 27.5 14.2 18.1 8.7 6.7 4.9
*Given most rural people did not have any health insurance before 2003, we assumed that total delivery expenditure was equal to the out-
of-pocket expenditure in 2002.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
1214 2010 Blackwell Publishing Ltd
The logarithmic value of the out-of-pocket expenditure
of facility-based delivery in 2007 was analysed by linear
regression (Table 6). After adjusting for age, education,
ethnicity, income, delivery mode and level of health
facility, NCMS participation correlated with reduced out-
of-pocket expenditure (coefcient )1.14 P < 0.05). In
addition, older age and higher education was correlated
with higher out-of-pocket expenditure. The out-of-pocket
expenditure increased as the income increased. The out-of-
pocket expenditure for caesarean section was higher than
for vaginal delivery. The expenditure in county level was
higher than in township-level health facility.
Discussion
In Chinas less developed western rural areas, the use of
facility-based delivery greatly improved, while the use of
pre-natal care showed a modest increase. There was
income-related inequality in using pre-natal and delivery
care, but differences between the income groups narrowed
in the period under study. The out-of-pocket expenditure
for facility-based delivery as a percentage of the annual
household income remained high in the low-income group,
although NCMS participation reduced the out-of-pocket
expenditure for facility-based delivery on average.
The data for this study are regionally representative. The
sample size is large, giving the analyses good statistical
power. However, there are some limitations. First, data on
the frequency and timing of pre-natal visits and the
expenditure of delivery care are subject to recall bias,
although a 1-year recall is unlikely to cause serious bias.
Secondly, the NCMS schemes varied across counties in
terms of the level of reimbursement and services covered,
so without data on these variations, the results should be
viewed as preliminary.
Maternal health care has been advocated as a main
strategy to prevent women from death and morbidity from
pregnancy-related causes (Filippi et al. 2006). Our ndings
show a notable increase in facility-based delivery and a
modest increase in utilising pre-natal care in western rural
areas. Consistent with other studies in rural China and
other developing countries, we found that lower education,
minority status, and higher parity were negatively related
to pre-natal and delivery care utilisation (Anson 2004;
Short & Zhang 2004; Simkhada et al. 2007; Sepehri et al.
2008). In addition, economic factors had a signicant
association with maternal health care utilisation. The
impressive increase in facility-based delivery may be partly
because of the reduced nancial burden on households,
while contrastingly, the willingness to pay for pre-natal
care is still low.
Financial difculty is usually considered one of the main
barriers to maternal health care utilisation in many low-
and middle-income countries (Smith & Sulzbach 2006;
Witter et al. 2007). Our study found that low-income
women were less likely to utilise pre-natal care and facility-
based delivery compared to medium- and high-income
women. From the mid-1980s, Chinese health sector
reforms shrunk the government allocation for health care.
Thereafter, the health care system became heavily depen-
dent on fee-for-service nancing. This combined with
prot-related bonus payments for health providers resulted
in a rapid increase in medical expenditures (Gao et al.
2002). High expenditure may have created a barrier to
poor women using maternal health care, particularly in
rural areas where demand is low. Bogg and colleagues have
reported similar ndings in other areas of rural China
(Bogg et al. 2002). Nevertheless, in our study, the differ-
ences by income in utilisation, especially of facility-based
delivery, lessened between 2002 and 2007. A reason to this
reduced difference can be the smaller increase in total and
out-of-pocket delivery expenditures in the low-income
group than the high-income group.
The NCMS covered a modest part of the expenditures
for facility-based deliveries. To our knowledge, no previ-
ous studies have reported the impact of the NCMS on the
expenditure of maternal health care. The NCMS may be a
useful way to share the nancial risk related to delivery and
to protect poor women from potential catastrophic pay-
ments. A study that estimated the impact of the NCMS on
health care in general using panel data collected in fteen
rural counties in 2003 and 2005 showed that on average
Table 6 Coefcients from logarithmic-linear model of out-of-
pocket expenditure for facility-based delivery, 2007
Coefcient
estimate
Standard
error P value
NCMS participation (No) )1.14 0.57 <0.05
Age (1524)
2529 0.51 0.34 0.14
3049 0.78 0.34 <0.05
Education (illiterate)
Primary school 0.30 0.52 0.57
Secondary school and above 1.07 0.51 <0.05
Ethnicity (minority)
Han )0.57 0.30 0.05
Income (Low income)
Medium income 0.44 0.35 0.21
High income 1.60 0.37 <0.01
Delivery mode (vaginal delivery)
Caesarean section 1.86 0.33 <0.01
Level of health facility (township level)
County level 2.26 0.30 <0.01
The reference group is shown in parentheses.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
2010 Blackwell Publishing Ltd 1215
the NCMS increased the out-of-pocket health expenditures
among all income groups, but reduced the incidence of
catastrophic spending among the poorest group (Wagstaff
et al. 2007).
During the study period, the Chinese government funded
a large maternal project: Reducing maternal mortality and
eliminating tetanus infection of newborn babies. It had a
wide coverage in western rural China and implemented
comprehensive intervention measures, such as giving health
education to rural women and a nancial subsidy to cover
part of facility-based delivery expenditures (Guo et al.
2008). Hence, it is difcult to separate the contributions of
the NCMS from those of that other project. Nevertheless,
this project was only for a certain period of time, and it,
like other short-term project, cannot sustainably solve the
nancing problems of maternal health care. In this context,
expanding NCMS can be a good opportunity to continue
to improve maternal health in rural China.
In 2007, the out-of-pocket expenditure of facility-based
delivery was about 14% of the annual household income
among the low-income group, which remains a large
payment at one time referring to the denition of
catastrophic spending that threshold varied from 10% to
40% of annual household income (Ranson 2002; Xu et al.
2003). No studies have reported how poor households in
rural China manage such high delivery expenditure.
Borrowing money and selling assets to cover delivery-
related expenditures have been reported in some African
and South Asian countries (Nahar & Costello 1998;
Storeng et al. 2008). In addition, in facility-based delivery,
there are much indirect expenditures, such as transporta-
tion and lodging for the accompanying family members. In
Tanzania and Nepal, these indirect expenditures have been
estimated to be 50% of the total normal delivery expen-
diture (Borghi et al. 2006).
In our study, income was positively related to having
pre-natal care and to the number of visits. Having pre-natal
care and the number of visits were associated with giving
birth at health facility. Pre-natal contacts are a good way to
give health education and psychosocial services to women,
and to raise womens awareness of safe delivery care
(Simkhada et al. 2007). However, ability and willingness
to pay can limit the use of pre-natal care. Studies in the
Philippines, Vietnam and some African countries showed
that health insurance coverage increased the probability of
having the recommended number of pre-natal visits (Smith
& Sulzbach 2006; Sepehri et al. 2008; Kozhimannil et al.
2009).
The new round of health-system innovations set up by
the Chinese government in 2009 set a goal of gradually
achieving an equalisation in basic public health care, with
maternal health care as one of the targets (National
Development and Reform Commission, 2009). We suggest
increasing the NCMS reimbursement levels for facility-
based delivery to further reduce the nancial burden for the
rural poor households. In addition, pre-natal care should
be included into the benet package of the NCMS to
encourage women to use it. However, the norms for the
adequate number of visits and care contents should be
revisited to correspond to international recommendations
(Villar et al. 2001).
Acknowledgements
The CHSI of Ministry of Health funded data collection and
management. Data analysis and interpretation was sup-
ported by a grant from the China Medical Board (CMB).
We thank the CMB for funding a training workshop in
data analysis. This study is part of the output of the
CHIMACA project (015396) funded by the European
Commission INCO Programme and co-ordinated by the
National Institute for Health and Welfare, Helsinki. We
thank Mark Phillips for polishing the English language of
the text. We appreciate the reviewers giving valuable
comments. QL appreciates the China Scholarship Council
for supporting her study abroad.
References
Adam T, Lim SS, Mehta S et al. (2005) Cost effectiveness analysis
of strategies for maternal and neonatal health in developing
countries. BMJ 331, 1107.
Anson O (2004) Utilization of maternal care in rural HeBei
Province, the Peoples Republic of China: individual and struc-
tural characteristics. Health Policy 70, 197206.
Bogg L, Wang K & Diwan V (2002) Chinese maternal health in
adjustment: claim for life. Reproductive Health Matters 10,
95107.
Borghi J, Ensor T, Somanathan A et al. (2006) Mobilising nancial
resources for maternal health. Lancet 368, 14571465.
Carrin G, Ron A, Yang H et al. (1999) The reform of the rural
cooperative medical system in the Peoples Republic of China:
interim experience in 14 pilot counties. Social Science &
Medicine 48, 961972.
Ekman B, Pathmanathan I & Liljestrand J (2008) Integrating
health interventions for women, newborn babies, and children: a
framework for action. Lancet 372, 9901000.
Feng X, Tang S, Bloom G, Segall M & Gu Y (1995) Cooperative
medical schemes in contemporary rural China. Social Science
&Medicine 41, 11111118.
Filippi V, Ronsmans C, Campbell OM et al. (2006) Maternal
health in poor countries: the broader context and a call for
action. Lancet 368, 15251541.
Gao J, Qian J, Tang S, Eriksson BO & Blas E (2002) Health equity
in transition from planned to market economy in China. Health
Policy Plan 17(Suppl.), 2029.
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
1216 2010 Blackwell Publishing Ltd
Guo Y, Zakus D & Liang H (2008) China: policy and practice of
MCH since the early 1990s. Maternal and Child Health Journal
12, 139148.
Jing F & Kaufman J (2008) Reproductive health in China: improve
the means to the end. Lancet 372, 16191620.
Kaufman J & Jing F (2002) Privatization of health services and the
reproductive health of rural Chinese women. Reproductive
Health Matters 10, 108116.
Kozhimannil KB, Valera MR, Adams AS & Degnan DR (2009)
The population-level impacts of a national health insurance
program and franchise midwife clinics on achievement of
pre-natal and delivery care standards in the Philippines. Health
Policy 92, 5564.
Li J (2004) Gender inequality, family planning, and maternal and
child care in a rural Chinese county. Social Science & Medicine
59, 695708.
Liu Y, Rao K, Wu J & Gakidou E (2008) Chinas health system
performance. Lancet 372, 19141923.
Ministry of Health (2004) Report of National Health Services
Survey 2003. Peking Union Medical College Press, Beijing (in
Chinese).
Nahar S & Costello A (1998) The hidden cost of free maternity
care in Dhaka Bangladesh. Health Policy and Planning, 13,
417422.
National Development and Reform Commission (2009) http://
www.china.org.cn/government/scio-press-conferences/2009-04/
09/content_17575401.htm (Access on 12 March 2010).
Ranson MK (2002) Reduction of catastrophic health care expen-
ditures by a community-based health insurance scheme in
Gujarat, India: current experiences and challenges. Bulletin of
the World Health Organization 80, 613621.
Sepehri A, Sarma S, Simpson W & Moshiri S (2008) How
important are individual, household and commune characteris-
tics in explaining utilization of maternal health services in
Vietnam? Social Science & Medicine 67, 10091117.
Short S & Zhang F (2004) Use of maternal health services in rural
China. Population Studies 58, 319.
Simkhada B, Teijlingen ER, Porter M & Simkhada P (2007) Fac-
tors affecting the utilization of antenatal care in developing
countries: systematic review of the literature. Journal of
Advanced Nursing 61, 244260.
Smith KV & Sulzbach S (2006) Community-based health insur-
ance and access to maternal health services: evidence from three
West African countries. Social Science &Medicine 66, 2460
2473.
Storeng KT, Baggaley RF, Ganaba R et al. (2008) Paying the price:
the cost and consequences of emergency obstetric care in
Burkina Faso. Social Science and Medicine, 66, 545557.
Villar J, Carroli G, Khan-Neelofur D, Piaggio GGP & Gulmezoglu
AM (2001) Patterns of routine antenatal care for low-risk
pregnancy. Cochrane Database of Systematic Reviews,
CD000934.
Wagstaff A, Lindelow M, Gao J, Xu L & Qian J (2007) Extending
health insurance to the rural population: an impact evaluation of
Chinas New Cooperative Medical Scheme. World Bank Policy
Resaerch Working Paper WPS 4150.
Wang Y (2007) Development of the New Rural Cooperative
Medical System in China. China & World Economy 15, 6677.
Wang H, Gu D & Dupre ME (2008) Factors associated with en-
rolment, satisfaction and sustainability of the New Cooperative
Medical Scheme program in six study areas in rural Beijing.
Health Policy 85, 3244.
Witter S, Arhinful DK, Kusi A & Zakariah-Akoto S (2007) The
experience of Ghana in implementation a user fee exemption
policy to provide free delivery care. Reproductive Health Mat-
ters 15, 6171.
Wu Z, Vlisaninen K, Li X & Hemminki E (2008) Maternal care in
rural China: a case study from Anhui province. BMC Health
Services Research 8, 55.
Xu K, Evans DB, Kawabata K et al. (2003) Household cata-
strophic health expenditure: a multicountry analysis. The Lancet
362, 111117.
You X & Kobayashi Y (2008) The new cooperative medical
scheme in China. Health Policy 91, 19.
Zhang L, Wang H, Wang L & Hsiao W (2006) Social capital and
farmers willingness-to-join a newly established community-
based health insurance in rural China. Health Policy 76,
233242.
Corresponding Author Qian Long, School of Public Health, Chongqing Medical University, No. 1 Yixueyuan Road, Chongqing,
China. E-mail: [email protected]
Tropical Medicine and International Health volume 15 no 10 pp 12101217 october 2010
Q. Long et al. Use of maternal health care in China
2010 Blackwell Publishing Ltd 1217
Copyright of Tropical Medicine & International Health is the property of Wiley-Blackwell and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.