Ten years of China’s new healthcare reform: a longitudinal study on changes in health resources

Abstract

Background
China launched a new round of healthcare-system reform in 2009 and proposed the goal of equal and guaranteed essential medical and health services for all by 2020. We aimed to investigate the changes in China’s health resources over the past ten years after the healthcare reform.

Methods
Data were collected from the China Statistical Yearbook and China Health Statistics Yearbook from 2009 to 2018. Four categories and ten indicators of health resources were analyzed. A descriptive analysis was used to present the overall condition. The Health Resource Density Index was applied to showcase health-resource distribution in demographic and geographic dimensions. The global and local Moran’s I were used to assess the spatial autocorrelation of health resources. Concentration Index (CI) was used to quantify the equity of health-resource distribution. A Geo-Detector model and Geographic Weighted Regression (GWR) were applied to assess the association between gross domestic product (GDP) per capita and health resources.

Results
Health resources have increased over the past ten years. The global and local Moran’s I suggested spatial aggregation in the distribution of health resources. Hospital beds were concentrated in wealthier areas, but this inequity decreased yearly (from CI=0.0587 in 2009 to CI=0.0021 in 2018). Primary medical and health institutions (PMHI) and their beds were concentrated in poorer areas (CI remained negative). Healthcare employees were concentrated in wealthier areas (CI remained positive). In 2017, the q-statistics indicated that the explanatory power of GDP per capita to beds, health personnel, and health expenditure was 40.7%, 50.3%, and 42.5%, respectively. The coefficients of GWR remained positive with statistical significance, indicating the positive association between GDP per capita and health resources.

Conclusions
From 2009 to 2018, the total amount of health resources in China has increased substantially. Spatial aggregation existed in the health-resources distribution. Health resources tended to be concentrated in wealthier areas. When allocating health resources, the governments should take economic factors into account.

Full text

Chenetal. BMC Public Health (2021) 21:2272
https://doi.org/10.1186/s12889-021-12248-9
RESEARCH
Ten years ofChina’s new healthcare reform:
alongitudinal study onchanges inhealth
resources
Jiang Chen1†, Zhuochen Lin1†, Li‑an Li1, Jing Li1, Yuyao Wang1, Yu Pan1, Jie Yang1, Chuncong Xu1,
Xiaojing Zeng1, Xiaoxu Xie2*† and Liangcheng Xiao1*†
Abstract
Background: China launched a new round of healthcare‑system reform in 2009 and proposed the goal of equal
and guaranteed essential medical and health services for all by 2020. We aimed to investigate the changes in China’s
health resources over the past ten years after the healthcare reform.
Methods: Data were collected from the China Statistical Yearbook and China Health Statistics Yearbook from 2009 to
2018. Four categories and ten indicators of health resources were analyzed. A descriptive analysis was used to present
the overall condition. The Health Resource Density Index was applied to showcase health‑resource distribution in
demographic and geographic dimensions. The global and local Moran’s I were used to assess the spatial autocorrela‑
tion of health resources. Concentration Index (CI) was used to quantify the equity of health‑resource distribution. A
Geo‑Detector model and Geographic Weighted Regression (GWR) were applied to assess the association between
gross domestic product (GDP) per capita and health resources.
Results: Health resources have increased over the past ten years. The global and local Moran’s I suggested spatial
aggregation in the distribution of health resources. Hospital beds were concentrated in wealthier areas, but this ineq‑
uity decreased yearly (from CI=0.0587 in 2009 to CI=0.0021 in 2018). Primary medical and health institutions (PMHI)
and their beds were concentrated in poorer areas (CI remained negative). Healthcare employees were concentrated
in wealthier areas (CI remained positive). In 2017, the q‑statistics indicated that the explanatory power of GDP per
capita to beds, health personnel, and health expenditure was 40.7%, 50.3%, and 42.5%, respectively. The coefficients
of GWR remained positive with statistical significance, indicating the positive association between GDP per capita and
health resources.
Conclusions: From 2009 to 2018, the total amount of health resources in China has increased substantially. Spatial
aggregation existed in the health‑resources distribution. Health resources tended to be concentrated in wealthier
areas. When allocating health resources, the governments should take economic factors into account.
Keywords: Concentration index, Geo‑Detector model, Geographic weighted regression, Moran’s I, Healthcare reform,
Health resources
© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco
mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Background
According to the World Health Organization, equity
is one of the most basic ethical principles of health-
resource allocation [1]. Health-resource allocation
reflects the distribution and flow of health resources in
Open Access
*Correspondence: xiexiaoxu@aliyun.com; xiaolch@mail.sysu.edu.cn
†Jiang Chen, Zhuochen Lin, Xiaoxu Xie and Liangcheng Xiao contributed
equally to this work.
1 Department of Medical Affairs, The First Affiliated Hospital, Sun Yat‑sen
University, Guangzhou, China
2 School of Public Health, Fujian Medical University, Fuzhou, China
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 2 of 13
Chenetal. BMC Public Health (2021) 21:2272
different regions. It is often used to measure the degree of
health equity, which is regarded as an important part of
social equity [2, 3]. Before the new healthcare reform, one
of the most considerable problems the Chinese medical
and healthcare system faced was the deficient and unbal-
anced distribution of health resources. e demographic
and geographic maldistribution of healthcare resources is
considered a prominent healthcare issue in China [4–6].
Studies have shown that inequality in the allocation of
health resources is closely associated with increased dis-
parities in health outcomes [7–9]. Frequently cited con-
sequences of the unequal health-resource allocation are
catastrophic health expenditures and impoverishment
[10]. Focus on the unfair distribution of health resources
is required to reduce health inequities [11]. In April
2009, China launched a new round of healthcare-system
reform and set out to improve equitable access to medi-
cal services [12]. From 2009 to 2011, the government
focused on increasing financial investment in the health
sector to expand insurance coverage and build infra-
structure. Since 2012, more emphasis has been placed
on healthcare-delivery reform to increase health-service
efficiency [10]. During this period, the government has
issued a series of policies to promote equitable access to
health resources, such as universal health insurance pro-
grams, zero-markup drug policy, patient-referral policy,
medical alliance policy. In particular, the Healthy China
2030 program of the Chinese government that advo-
cates to “accelerate the expansion of high-quality health
resources and the balanced distribution of such resources
among different regions” [13] has been regarded as a
breakthrough for improving health [14].
Ten years after the new healthcare reform, the medi-
cal insurance system with almost universal coverage
has been established [15], and the accessibility of medi-
cal services has been sharply improved [10]. Has a com-
mensurate improvement in health-resource settings
occurred? Has the allocation of health resources been
gradually equal?
Previous studies have been conducted to assess health-
resource allocation in China with different priorities.
Such as emergency health resources [16, 17], primary
medical and health institutions(PMHI) [18], comparison
of the health resources and medical-service allocation
between hospitals and PMHI [9], hospital beds [19], and
health-workforce distribution [5]. e official description
of health resources is usually divided into four categories,
namely, healthcare institutions, beds, health personnel,
and health expenditure [20]. In previous studies, they
have often been cited as a proxy indicator of health
resources, respectively [5, 9, 16–19]. To our knowledge,
few scholars have conducted a systematic survey of Chi-
na’s health resources within a span of ten years after the
new healthcare reform so far. In particular, the involve-
ment of health expenditure is scarce. In our study, we
included the four categories of health resources. rough
the research and analysis from the combination of these
four aspects, we can have a more systematic and compre-
hensive understanding of China’s health resources. When
assessing health-resource allocation, the results based on
population density and geographic space showed a differ-
ent situation [3, 9, 16, 17, 19]. Geographic inequalities in
access to health resources have been addressed by previ-
ous studies as a challenging issue for many countries [4,
21–23]. e improvement of equity in spatial access to
health resources is expected to potentially enhance the
effectiveness of healthcare service utilization [4, 24, 25].
us, when describing the health-resource allocation, we
compared it demographically and geographically, respec-
tively. Measurements of health-resource distribution
based on population density were more common, but it
may ignore the effects of geography. e Health Resource
Density Index (HRDI) was another tool that can be used.
Compared with the assessment based on population den-
sity, it can mediate bias and influences owing to a single
aspect of population or geographic area [9, 26]. Moran’s
I was often applied to assess spatial autocorrelation. It
included two types: Global Moran’s I and Local Moran’s I.
e former was used to measure the general spatial auto-
correlation and the spatial distribution of the research
object, and the latter measured the local spatial autocor-
relation and the cluster regions [27–29].
Many methods and indicators are used to study the
equity of health-resource allocation, such as Concentra-
tion Index (CI), Lorenz Curve and Gini Coefficient, and
eil Index [30]. Each method and index have its own
advantages and disadvantages, and the applicable con-
ditions are not the same. Among these equity research
methods, CI is extensively used to measure the degree
of equity in the allocation of health resources associated
with socioeconomic conditions [30, 31]. When measur-
ing the equity of health-resource allocation, CI consid-
ers the social and economic conditions. It can accurately
reflect the allocation of health resources to different
social strata [32], which can help the government under-
stand the allocation of health resources at all levels of
society.
Gross domestic product (GDP) per capita is considered
when calculating CI, and it has been addressed by previ-
ous studies as a factor affecting health-resource distribu-
tion [33, 34]. A Geo-Detector model, which is often used
to measure similarities in the spatial distributions of two
variables [35–37], has been developed to evaluate the
spatial and temporal matching levels between GDP per
capita and health resources at the provincial level. Geo-
graphic weighted regression (GWR) is a local regression
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 3 of 13
Chenetal. BMC Public Health (2021) 21:2272
based on geospatial weighting, which can effectively cap-
ture the impact of independent variables on outcomes in
local areas [38]. It has been applied to assess the impact
of GDP per capita on the distribution of health resources.
e present longitudinal study aimed to systemati-
cally investigate the changes in China’s health resources
(including overall condition, spatiotemporal distribution,
equity of allocation, and the association between GDP
per capita and health resources) over the past ten years
after the healthcare reform. Our findings are expected
to comprehensively show the overview of changes in the
distribution of health resources in China, and the results
can be applied to provide implications on future spatial
allocation of health resources and evidence-based health
planning procedures.
Methods
Data sources
e indicators of health resources were adopted in
accordance with the " Statistical Communique on the
Development of China’s Health Undertakings " published
by the China National Health Commission every year.
ese indicators are commonly cited in studies on health
resources in China [5, 9, 16–19]. Hospitals and PMHI are
the main places for diagnosis and treatment, and they
account for the vast majority proportion of medical and
healthcare institutions (more than 97% in 2018 [20]).
us, they were included as indicators of healthcare insti-
tutions in the study. Correspondingly, their beds were also
included. Health personnel refers to employees engaged
in the healthcare institutions. In this study, licensed doc-
tors (LD), registered nurses (RN), and healthcare employ-
ees (HCE) were included. Total health expenditure (THE)
is usually divided into three types, including government
health expenditure (GHE), social health expenditure
(SHE), and Out-of-pocket payments (OOPs). Data for
demographic, geographical, and socioeconomic includ-
ing total population, geographical area, and GDP per
capita were also included for the distribution assessment
of health resources. ese indicators are defined in the
China Statistical Yearbook and China Health Statistical
Yearbook and are listed in Additional file1.
e data related to these indicators of 31 provinces,
autonomous regions, and municipalities directly under
the central government originate from China Statistical
Yearbook and China Health Statistics Yearbook from the
year 2009 to 2018. Due to inconsistent statistical stand-
ards, Hong Kong, Macau, and Taiwan were not included
in this study.
Study design
is study was divided into three steps.
Description ofthegeneral situation
For the first step, we described the overall condition of
health resources in China over the past ten years from
four categories. Ten indicatorsincluding hospital, PMHI,
hospital beds, PMHI beds, HCE, LD, RN, GHE, SHE, and
OOPs were studied. In addition to the above ten indica-
tors, we also introduced the calculation of Beds per 1,000
people, HCE per 1,000 people, GHE Per Capital, the pro-
portion of OOPs in THE, and the proportion of THE in
GDP for a overview of health resources.
Demographic andgeographic distribution ofhealth
resources
For the second step, we explored the demographic and
geographic distribution of health resources. Since we
introduced calculation methods of per capita or per thou-
sand people in the comparison process, this was not appli-
cable to healthcare institutions in practice. erefore, in
this process, we mainly compared three categories of indi-
cators, namely beds, health personnel, and health expend-
iture. We used maps to visually present the distribution of
beds (both in hospitals and PMHI per 1,000 people), HCE
per 1,000 people, and THE per capita (refers to the ratio
of THE in a year to the average population) between 2009
and 2018. We also included HRDI to present the inte-
grated health-resource distribution from the aspects of
population density and geographic area. HRDI was calcu-
lated as the geometric mean of health resources per 1,000
people and per square kilometers. e following formula
was used to calculate HRDI [9, 26]:
Yi represents the health resource of unit I; Pi represents
the population of unit I, and Ai represents the area of
unit I.
e results of HRDI were also presented by maps. To
measure the global autocorrelation, we introduced the
global Moran’s I index. e global Moran’s I is an impor-
tant index to measure spatial autocorrelation with the
range of -1 to 1 [39]. If Moran’s I is larger than 0, the
resources had spatial disparity, indicating that a larger
(smaller) resource corresponded with easier (harder)
aggregation. If Moran’s I was smaller than 0, the resources
had spatial heterogeneity, indicating that a larger (smaller)
resource corresponded with less likelihood of aggregation.
When Moran’ I was 0 (or P > 0.05), resources were ran-
domly distributed in space and had no spatial correlation.
When the global Moran’s I is statistically significant, the
local Moran’s I can be further analyzed. e local Moran’s
I can analyze whether an indicator in a local area has spa-
tial correlation, which can be divided into two parts: (1)
HRDI =(Yi/Pi)(Yi/Ai)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 4 of 13
Chenetal. BMC Public Health (2021) 21:2272
e level of an indicator in the region compared with the
overall; (2) e level of indicators in the surrounding areas
compared with the overall. e global Moran’s I and local
Moran’s I can be calculated by:
Where n is the number of the province, yi and yj
are the resource of province i and j, respectively,
y
is the mean of the resource of all provinces,
S
0=
n
i=
1
n
j=
1ω
ij
, and
ωij
is the spatial weight value
of province i and j calculated from the spatial dis-
tance using Euclidean distance. It should be noted
that the data points used for the calculation of
ωij
were different between global Moran’s I and local
Moran’s I.
GobalMoran
’sI =
n
S0
*
n
i=1
n
j=1ωij(yi−y)(yj−y
)

n
i=1
(
yi
−
y
)2
LocalMoran
’sI =Ii=(yi−y)

n
k=1
y
i
−y

2∕(n−1)
*
n

j≠i
wij(yj−y
)
Assessment oftheequity andtheassociation
betweeneconomic factor andhealth resources
For the third step, we used CI to quantify the degree of inequal-
ity in the allocation of health resources. We introduced a Geo-
Detector model and GWR to further measure the association
between GDP per capita and health-resource distribution.
CI is recognized as a superior tool to measure the equal-
ity of health-resource allocation associated with socio-
economic status [32]. e following figure was used to
demonstrate the definition of CI.
e x-axis is the cumulative share of the population,
ordered by GDP per capita from lowest to highest, and the
y-axis represents the cumulative share of health resources.
e concentration curve presents the cumulative share
of the health resources against the cumulative share of
the population. A in Fig.1 is the area between the line of
equality (the 45° line) and the concentration curve. S is the
area under the concentration curve. e CI is calculated as
twice the area A. e following exact computational for-
mula was used to calculate the CI.
S
=1
2
n−1
∑
i=0
(Yi+Yi+1)(Xi+1−Xi
)
CI
=
2
∗(
0.5
−
S
)
Fig. 1 Graphical definition of CI
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 5 of 13
Chenetal. BMC Public Health (2021) 21:2272
e CI is bound from -1 to +1. Where 0 means com-
plete equity, and ±1 means complete inequity. A negative
value indicates health resources are disproportionately
concentrated among the poorer, whereas a positive value
means that health resources are disproportionately con-
centrated within wealthier populations.
A Geo-Detector model is used to measure structured
spatial heterogeneity of health resources and determine
whether the health resources within strata are more simi-
lar than that between strata. Given the q-statistic calcu-
lated by the Geo-Detector model, we can say that 100q%
information of health resources can be explained by
structured heterogeneity based on the independent vari-
able (the GDP per capita here). e q-statistic has a range
of 0 to 1 and can be calculated as follows [36, 37]:
Where L is the number of strata of GDP per capita,
Nh
and N is the number of provinces in strata h and the num-
ber of all provinces, respectively, and
σ2
h
and
σ2
are the
variance of resources in strata h and all data.
L
h=
1N
h
σ
2
h
is within sum of square and
Nσ2
is total sum of squares.
Our research divided GDP per capita into four grades
using lower quartile, median, and upper quartile (L=4).
However, the calculation of q-statistics needs to discre-
tize the independent variables into the qualitative vari-
able, which may cause the loss of information. erefore,
we used GWR to analyze the impact of economic fac-
tors on the distribution of health resources. In our study,
Gaussian kernel was used, and the optimal bandwidth
was selected using leave-one-out cross validation.
R studio software (version 4.1.1) was used to per-
form all analyses. P < 0.05 was considered statistically
significant.
Results
Overview ofhealth resources overthepast ten years
Changes in the number of health resources in China
from 2009 to 2018 are shown in Table1. Generally, health
resources, including healthcare institutions, beds, health
personnel, and health expenditures have increased over
the past ten years. Important indicators for the allocation
of resources in the global study of health-service systems
such as beds per 1,000 people and HCE per 1,000 people
show a steady growth trend. Compared with 2009, num-
bers in 2018 increased by 83.86% and 51.11%, respectively.
e health investment of the Chinese government is also
growing. Over the past decade, GHE has grown steadily
at an annual growth rate of 11.88%. Correspondingly, the
GHE per capita in 2018 increased by 2.6 times compared
q
=1−
L
h=1Nhσ
2
h
Nσ
2
with 2009. e structure of health expenditure has also
improved. e proportion of OOPs in THE continued
to decrease, i.e., from 37.46% to 2009 to 28.61% in 2018,
resulting in the lowest level in nearly ten years. Meanwhile,
the proportion of SHE in THE has increased from 35.08%
to 2009 to 43.66% in 2018.
Spatiotemporal distribution ofhealth resources
In general, the health resources (beds per 1,000 people,
THE per capita, and HCE per 1,000 people) shown in
Fig. 2 A-F indicated a remarkable increase, respectively.
However, health-resource disparity existed among differ-
ent regions in both years. Economically developed areas
such as Beijing, Shanghai, Tianjin, Jiangsu, and Zhejiang
were wealthier in health resources than other places. Some
exceptions existed. Although Xinjiang was relatively back-
ward in economic development (GDP per capita ranked
21st and 19th out of 31 provincial areas in 2009 and 2018,
respectively), it had the highest number of beds per 1,000
people in China in 2009 and 2018 (Fig.2-A and -D). e
results of HRDI are shown in Fig.2G-L. After adjusting for
the geographical area, although disparities in the distribu-
tion of health resources remained, the distribution situa-
tion changed. Developed areas such as Beijing, Shanghai,
Tianjin, Jiangsu, and Zhejiang still had higher values. How-
ever, in regions with large geographical areas such as Xinji-
ang and Tibet, their advantages in the distribution of health
resources no longer exist. e results of each indicator are
shown in Additional file2A-C.
Based on the calculation results of the global Moran’s I
index, we can see a spatial disparity in the distribution of
the health resources from 2009 to 2018 (Moran’s I >0 and
P<0.05, Table2). is finding was consistent with the con-
clusion of uneven distribution of HRDI shown in Fig.2G-L.
e results of local Moran’s I are shown in Table3. Due to
space constraints, Table3 only shows the results of some
provinces; please refer to Additional file 3 for detailed
results. It can be seen that Beijing, Shanghai, and Tianjin
had spatial correlation in most years, and the central and
surrounding areas were both higher than the overall level
(H, H). e local Moran’s I in Jiangxi, Jiangsu, Zhejiang,
Xinjiang, and Tibet had statistical significance in some
years. e first three provinces were (H, H) (like Beijing
and Shanghai). Xinjiang and Tibet were (L, L), indicating
that the indicators of the two provinces themselves and
surrounding provinces were lower than the average.
Equity intheallocation ofhealth resources
andtheassociation betweenGDP percapita andhealth
resources
e description of CI comprised four parts: healthcare
institutions, beds, health personnel, and health expendi-
ture. e trend of each indicator is shown in Fig. 3.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 6 of 13
Chenetal. BMC Public Health (2021) 21:2272
Table 1 Overall condition of health resources in China from 2009 to 2018
Abbreviations:PMHIPrimary Medical and Health Institutions, GHEGovernment Health Expenditure, SHESocial Health Expenditure, OOPsOut-of-Pocket Payments, THETotal Health Expenditure, LDLicensed doctors,
RNRegistered Nurses, HCEHealth Care Employees;aFigures adjusted for ination and calculated in Chinese yuan in 2018 prices
Institutions Beds Health Personnel Health Expenditure
Year Hospitals PMHI Hospitals PMHI Beds per
1,000
people
LD RN HCE HCE per
1,000
people
GHEa(100
million
yuan)
GHE per
captialaSHEa(100
million
yuan)
OOPsa(100
million
yuan)
OOPs(%
of THE) THE(% of
GDP)
2009 20,291 882,153 3,120,773 1,099,662 3.16 1,905,436 1,854,818 7,781,448 5.83 5972.16 447.52 7631.57 8148.32 37.46% 5.03%
2010 20,918 901,709 3,387,437 1,192,242 3.42 1,972,840 2,048,071 8,207,502 6.12 6993.64 521.56 8779.86 8602.18 35.29% 4.84%
2011 21,979 918,003 3,705,118 1,233,721 3.67 2,020,154 2,244,020 8,207,502 6.09 8733.09 648.17 9847.25 9917.11 34.80% 4.98%
2012 23,170 912,620 4,161,486 1,324,270 4.05 2,138,836 2,496,599 9,115,705 6.73 9528.14 703.68 11334.69 10911.20 34.34% 5.20%
2013 24,709 915,368 4,578,601 1,349,908 4.36 2,285,794 2,783,121 9,790,483 7.20 10500.39 771.68 12533.17 11812.98 33.90% 5.32%
2014 25,860 917,335 4,961,161 1,381,197 4.64 2,374,917 3,004,144 10,234,213 7.48 11319.78 827.58 14378.39 12087.70 31.99% 5.48%
2015 27,587 920,077 5,330,580 1,413,842 4.91 2,508,408 3,241,469 10,693,881 7.78 13223.80 962.00 17497.11 12713.15 29.27% 5.95%
2016 29,140 926,518 5,689,000 1,441,900 5.16 2,651,398 3,507,166 11,172,945 8.08 14466.72 1046.26 19860.55 13871.65 28.78% 6.23%
2017 31,056 933,024 6,120,500 1,528,500 5.50 2,828,999 3,804,021 11,748,972 8.45 15509.99 1115.76 22703.99 15434.73 28.77% 6.36%
2018 33,009 943,639 6,519,800 1,583,500 5.81 3,010,376 4,098,630 12,290,325 8.81 16399.13 1175.24 25810.78 16915.89 28.61% 6.57%
Annual
growth
rates
5.56% 0.75% 8.53% 4.13% 6.99% 5.21% 9.21% 5.21% 4.69% 11.88% 11.32% 14.50% 8.45% .. ..
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 7 of 13
Chenetal. BMC Public Health (2021) 21:2272
e distribution of hospitals was relatively fair, with
CI values close to 0(CImax= 0.0037, CImin= -0.0406).
PMHI had always been primarily distributed in poorer
areas, and the trend was an increasing one (from CI=-
0.0646 in 2009 to CI=-0.0847 in 2018) (Fig.3-A). Hos-
pital beds tended to be distributed in wealthier areas,
but this inequity decreased yearly (from CI=0.0587 in
2009 to CI=0.0021 in 2018). Beds in PMHI were primar-
ily distributed in poorer areas. From 2009 to 2013, this
kind of inequity expanded (from CI=-0.0581 in 2009 to
CI=-0.0955 in 2013). Since 2013, this trend has eased
(Fig. 3-B). Overall, the distribution of hospital beds
and PMHI beds tended to be fair gradually, CI values
tended to be close to 0. HCE was primarily distributed
in the wealthier areas, but the overall trend of CI values
was downward (from CI=0.0555 in 2009 to CI=0.0356
in 2018). LD (from CI=0.0909 in 2009 to CI=0.0719 in
2018) and RN (from CI=0.0881 in 2009 to CI=0.0526
in 2018) had the same trend, but their distribution was
more unfair (Fig. 3-C). Health expenditure, including
GHE, SHE, and OOPs were primarily concentrated in
wealthier areas, with relatively stable positive CI val-
ues. e inequity degree of SHE was the highest (from
CI=0.2496 in 2009 to CI=0.1964 in 2017) (Fig. 3-D).
Meanwhile, GHE had the lowest level of CI, indicating a
relatively fair distribution compared with the other two
types of health expenditure.
Fig. 2 Comparison of health‑resource distribution based on population density and HRDI
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 8 of 13
Chenetal. BMC Public Health (2021) 21:2272
After calculation, the q-statistics were around 0.29 to
0.51 with statistical significance expected for the beds in
2014 to 2016(the P values were 0.052, 0.062 and 0.062,
respectively; Table 4). From the perspective of verti-
cal years, the q-statistics of health resources increased,
with beds increasing from 0.363 to 2009 to 0.401 in
2018, health personnel increasing from 0.412 to 2009 to
0.510 in 2018, and health expenditure increasing from
0.378 in 2012 to 0.425 in 2017. e increasing trend of
q-statistics indicated that the matching degree between
Table 2 The global Moran’s I of HRDI for health resources in different years
*:P< 0.05; #:P< 0.01
HRDI 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Beds 0.047#0.0389*0.054#0.077#0.079#0.073#0.081#0.095#0.097#0.096#
Health personnel 0.055#0.051#0.070#0.077#0.080#0.077#0.083#0.105#0.105#0.103#
Health expenditure ‑ ‑ ‑ 0.071#0.073#0.078#0.079#0.065#0.057#‑
Table 3 The local Moran’s I of HRDI for health resources in different years
*:p< 0.05; **:p< 0.001
Area Beds Health personnel Health expenditure
2009 2018 2009 2018 2012 2017
Beijing 0.923*(H,H) 1.637 1.245*(H,H) 3.529**(H,H) 1.477**(H,H) 2.361*(H,H)
Jiangxi 1.355 ‑0.076 1.375 0.063 1.810*(H,H) 0.108
Jiangsu 0.179 1.777*(H,H) 0.18 1.003 0.221 0.627
Shanghai 0.371 4.250**(H,H) 0.564 3.171*(H,H) 5.894**(H,H) 1.905*(H,H)
Tianjin 1.755*(H,H) 1.643 2.358*(H,H) 3.467*(H,H) 1.282**(H,H) 2.651*(H,H)
Tibet 0.731 0.235*(L,L) 0.775 0.18 0.47 0.081
Xinjiang 0.784 0.914*(L,L) 0.726 0.655*(L,L) 0.207 0.26
Zhejiang 0.022 1.821*(H,H) 0.046 1.907*(H,H) 0.094 1.11
Fig. 3 Trends of CI for ten indicators
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 9 of 13
Chenetal. BMC Public Health (2021) 21:2272
the distribution of health resources and GDP per capita
was gradually improving. In a horizontal comparison
of the three health resources indicators, the q-statistics
of health personnel and health expenditure were higher
than that of beds every year, indicating that GDP per
capita had a more significant impact on the distribution
of health personnel and health expenditure than beds.
Table 5 shows the GWR results of health resources in
10 years. e table included the overall regression coef-
ficient, minimum and maximum values of the GWR coef-
ficient, and adjustment R2 of GWR. e results showed
that all the GWRs were statistically significant, and the
GWR coefficients in different areas in each GWR fluc-
tuated little. e minimum and maximum of the coef-
ficients were close to the overall regression coefficient.
e adjusted R2 of GWR mainly concentrated from 0.5 to
0.7, indicating that economic factors closely impacted on
health resources.
Discussion
is was a longitudinal study that assessed the changes in
health resources in China in the context of new health-
care reform for ten years. From 2009 to 2018, the total
amount of health resources in China showcased a steady
increase. Previous related studies focusing on the differ-
ent periods have similar results [3, 40]. A review study
about the ten years of healthcare reform in China has
shown that the annual growth rate of both THE per cap-
ita and GHE per capita is higher than that of GDP per
capita in the same timeframe [10]. is mismatch implies
that the growth in health resources is due to economic
growth and the increasing emphasis of Chinese govern-
ments on healthcare systems. In the early stages of the
reform, the government significantly increased financial
investment in the health sector. Liu etal. showed that
from 2009 to 2013, health investment had grown con-
siderably, indicating a twofold increase [3]. e level of
health investment in China has considerably improved,
Table 4 The q‑statistics of HRDI in different years
*: P< 0.05; #: P< 0.01
HRDI 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Beds 0.363*0.343*0.340*0.339*0.322*0.305 0.291 0.291 0.407*0.401*
Health personnel 0.412*0.389*0.387*0.387*0.365*0.337*0.362*0.366*0.503#0.510#
Health expenditure ‑ ‑ ‑ 0.378*0.376*0.371*0.384*0.378*0.425#‑
Table 5 The results of GWR of health resources in 10 years
*: p < 0.05; **: p < 0.001
Year Beds Health personnel Health expenditure
2009 0.08** 0.165** ‑
(0.062~0.085,0.669) (0.143~0.173,0.727)
2010 0.072** 0.146** ‑
(0.055~0.077,0.618) (0.127~0.153,0.685)
2011 0.062** 0.131** ‑
(0.063~0.073,0.54) (0.133~0.154,0.629)
2012 0.059** 0.127** 70829.639**
(0.059~0.065,0.473) (0.129~0.14,0.579) (69154.848~79726.965,0.573)
2013 0.057** 0.121** 74987.434**
(0.058~0.064,0.476) (0.123~0.146,0.577) (76728.511~87267.249,0.579)
2014 0.056** 0.12** 81311.573**
(0.057~0.066,0.482) (0.122~0.143,0.582) (83350.619~93382.926,0.613)
2015 0.056** 0.122** 89729.788**
(0.057~0.063,0.494) (0.124~0.141,0.611) (90274.719~100659.859,0.632)
2016 0.055** 0.119** 96768.578**
(0.05~0.055,0.545) (0.116~0.119,0.649) (94117.045~97879.254,0.645)
2017 0.055** 0.12** 100961.734**
(0.05~0.054,0.565) (0.118~0.121,0.701) (98550.737~102123.025,0.657)
2018 0.052** 0.119** -
(0.048~0.052,0.543) (0.117~0.12,0.707)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 10 of 13
Chenetal. BMC Public Health (2021) 21:2272
but the proportion of THE in GDP remains lower than
that in many countries [41]. Except for the financial
investment, the Chinese government gradually turns
its emphasis to healthcare-delivery reform [10]. Policies
such as zero-markup drug policy, patient-referral policy,
medical alliance policy, and reform of the medical-insur-
ance payment system have been adopted to promote the
reform. We found another important improvement was
the positive change in the structure of health expendi-
ture. e share of OOPs in THE has continued to decline,
indicating that the problems of catastrophic health
expenditures and impoverishment caused by limited
health resources were alleviated. e possible reasons
were as follows: First, China vigorously developed social
medical insurance and increased its coverage. Since 2013,
the coverage has remained above 95% [15]. Second, the
government has gradually expanded the medical services
covered by social medical insurance, reduced the coin-
surance ratio, and increased the maximum reimburse-
ments [42]. ird, the central government advocated the
reform of medical-insurance payment methods [43]. New
payment methods including global budgets, diagnosis-
related groups, and case-based payments were applied to
pilot experiments, which has been preliminarily proven
to be effective in reducing OOPs [10]. In order to achieve
the goal of reducing the share of OOPs in THE to 25% by
2030 [13], the Chinese government needs to continue to
deepen the reform, including but not limited to the above
aspects. It is worth noting that although the situation of
health resources has improved significantly, there has
also been an increase in the health needs of the people.
For example, the number of RN per 1,000 people peaked
at 2.94 in 2018, which is still far behind the target of 4.7
for Healthy China 2030 [13].
Maps were used to visually display the distribution of
health resources in each area throughout the research
period. Figure2A-F shows an obvious increase in health
resources, but the disparities between different regions
still existed in both years. is kind of difference was
measured based on population density. Large, sparsely
populated regions such as Tibet and Xinjiang had a
remarkable advantage. However, this finding was not
consistent with the fact that the National Health Com-
mission sent medical technicians to Xinjiang and Tibet
for counterpart aid almost every year. is distribution
of health resources did not match the economic level,
either. For example, the GDP per capita of Guangdong
and Fujian in 2018 was 86,412 Yuan, 91,197 Yuan, respec-
tively, higher than those of Qinghai (47,689) and Sichuan
(48,883) [44]. However, the number of beds per 1,000
people and the number of HCE per 1,000 people were
lower than those of Qinghai and Sichuan (in 2018, the
number of beds per 1,000 people was 4.3 in Guangdong,
4.6 in Fujian, 6.4 in Qinghai, and 7.0 in Sichuan; the num-
ber of HCE per 1,000 people was 8.1 in Guangdong, 8.1
in Fujian, 9.5 in Qinghai, and 8.9 in Sichuan). To better
integrate the influence of population density and geo-
graphical area, we adopted HRDI for further compari-
son. e HRDI for beds, health expenditure, and health
personnel in Shanghai and Beijing remained the first and
second, respectively, whereas Xinjiang almost remained
the third from the bottom. is finding differed from the
situation measured by population density, in which Xinji-
ang ranked at the top. Previous studies had similar results
[3, 9, 16, 17, 19]. Figure2G-L shows that the distribution
of health resources presented a general trend of gradual
enhancement from west to east. Health resources were
primarily concentrated in the economically developed
eastern regions. is kind of distribution had positive
spatial autocorrelation with statistical significance every
year using the global Moran’s I index. e local Moran’s
I results also indicated that there was the obvious spatial
aggregation of health resources in economically devel-
oped areas such as Beijing and Shanghai, whereas health
resources were scarce in Xinjiang, Tibet, and the poor
surrounding areas. is verified the necessity of health
poverty alleviation projects proposed by “Healthy China
2030” to increase support for the development of medi-
cal and health institutions in poor central and western
regions [13].
Overall, since the new healthcare reform, the uneven
distribution of health resources still existed, but the
degree of unfairness had gradually decreased. e CI for
hospitals gradually approached 0, indicating that the dis-
tribution of hospitals in various provincial regions was
gradually equal. CI for PMHI gradually increased in a
negative direction, meaning that they were primarily dis-
tributed in economically underdeveloped areas. is dis-
tribution may be reasonable as this was consistent with
their positioning. CI for beds (in hospital and PMHI)
gradually approached 0, indicating that the distribution
of beds in all provincial areas was gradually equitable.
Moreover, compared with the other three categories of
health-resource distribution, beds distribution showed
better fairness. is finding was consistent with previ-
ous research [3]. e THE directly indicates the level of
the country’s investment in the health field. However, the
distribution of health expenditure was the most unequal
among the four categories of health resources. SHE was
primarily distributed in wealthier areas, and it was more
unequal than the other two types of health expenditure.
is finding was partly due to the fact that the contribu-
tion amount and reimbursement ratio of social medi-
cal insurance in each provincial area differed. e more
affluent areas had greater health-financing intensity.
eir commercial medical insurance and private health
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 11 of 13
Chenetal. BMC Public Health (2021) 21:2272
resources were more abundant, and social donations
were more abundant. CI for OOPs was positive, indi-
cating that OOPs for health services were higher in the
relatively high-income groups. Liu et al. found that a
higher family income corresponded with a higher health
expenditure [45]. is finding was consistent with the
findings of our study. Health personnel tended to be dis-
tributed to wealthier places. e same conclusion has
been drawn in previous studies [9, 46]. Other countries
such as ailand also face the same problems [47]. e
Chinese government has been committed to addressing
the problem of unequal distribution of health personnel,
and measures including giving the primary medical tech-
nicians preference to professional title assessment, salary
reform, and education have been taken to improve the
equity [48]. e results can be reflected in the changes
in CI for health personnel that the level of unfairness is
gradually decreasing.
e q-statistics gathered from the Geo-Detector model
indicated a match between GDP per capita and health
resources. e growing trend of q-statistics indicated an
increased matching degree between GDP per capita and
health resources. In 2017, more than 50% of information
of health personnel and 40% information of beds and
health expenditure can be explained by GDP per capita
with statistical significance. is finding was consistent
with a previous study declaring that the GDP per capita
had spatially positive clustered impacts on bed distribu-
tion [34]. However, their matching degree was still rela-
tively low, with a maximum of about 50%, suggesting that
there were still other factors that may affect the distri-
bution of health resources. is study found that health
personnel and health expenditure were more affected by
GDP per capita than beds. Consequently, the government
should consider more economic factors when consider-
ing the balanced distribution of health personnel and
health expenditure. e results of GWR indicated that
the GDP per capita was positively associated with health-
resources distribution, which was complementary to the
Geo-Detector model results. e government should be
more alert about the result as the economic factor could
widen the gap in health inequality because of its clustered
impacts on health resources.
is study also has some limitations. In CI calculation,
health-expenditure data in a few provinces in 2009, 2010,
and 2011 were not published, and the health expenditure
of each province in 2018 was not yet available. Instead, we
selected the data in 2012 and 2017 to calculate CI for quali-
tative judgment. Missing data may have different effects on
the results. In addition, due to the lack of a price index in
the health sector, we cannot compare the prices of health
services by region, which may have a potentially uncertain
impact on equity comparisons. For another, we discussed
the health-resource distribution at the provincial level, and
the situation inside the province has not been revealed. In
addition, this paper only preliminarily explored the asso-
ciation of economic factors with the distribution of health
resources, and other factors that may affect the allocation
of health resources need to be further studied.
Conclusions
Since implementing the new healthcare reform, China’s
health resources have shown significant growth in the ten
years. e distribution of health resources showed spatial
aggregation. Overall, the distribution of health resources
tended to be equitable gradually. e distribution of beds
was more equitable than other health resources. PMHI and
their beds were concentrated mostly in poor areas. Health
expenditure and health personnel tended to be concen-
trated in wealthier areas. SHE had the highest inequity.
GDP per capita was positively associated with the spatially
clustered distribution of health resources. Economic fac-
tors should be considered as an important factor in the bal-
anced allocation of health resources. is study can provide
decision makers with an improved understanding of the
current situation of health-resource allocation. Equity and
accessibility should continue to be an important considera-
tion in the allocation of health resources by governments.
How to optimize the allocation of health resources requires
further and more comprehensive research.
Abbreviations
GDP: Gross Domestic Product; CI: Concentration Index; HRDI: Health Resource
Density Index; PMHI: Primary medical and health institutions; GHE: Govern‑
ment Health Expenditure; SHE: Social Health Expenditure; OOPs: Out‑of‑
Pocket Payments; THE: Total Health Expenditure; HCE: Health Care Employees;
LD: Licensed doctors; RN: Registered Nurses; GWR : Geographic Weighted
Regression.
Supplementary Information
The online version contains supplementary material available at https:// doi.
org/ 10. 1186/ s12889‑ 021‑ 12248‑9.
Additional le1.
Additional le2.
Additional le3.
Acknowledgements
We appreciate the National Health Commission for its yearbooks, which
provide the data for this study. We would also like to express our appreciation
to all participants in this study for their participation and cooperation.
Authors’ Contributions
JC, ZL, XX, and LX conceived the study, analyzed the data, and drafted the
manuscript. LL, JL, and YW made substantial contributions to the study design,
data analysis, and modification of manuscript. YP, JY, CX, and XZ contributed
to the data collection, literature review, data interpretation and revised the
manuscript. All authors have read and approved the final version.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 12 of 13
Chenetal. BMC Public Health (2021) 21:2272
Funding
No external funding was received in this research.
Availability of data and materials
Please contact corresponding authors for data requests.
Declarations
Ethics approval and consent to participate
Not applicable. This study only analyzed data from published secondary
sources and did not involve any specific human subjects.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Received: 29 March 2021 Accepted: 16 November 2021
References
1. Solar O, Irwin A. A Conceptual Framework for Action on the Social
Determinants of Health. Geneva: World Health Organization; 2010.
2. Culyer AJ, Wagstaff A. Equity and equality in health and health care. J
Health Economics. 1993;12(4):431–57.
3. Liu W, Liu Y, Twum P, Li S. National equity of health resource allocation
in China: data from 2009 to 2013. Int J Equity Health. 2016;15:68.
4. Zhang Y, Yang H, Pan J. Gaining from rational health planning: Spatial
reallocation of top‑tier general hospital beds in China. Comput Indus‑
trial Eng. 2021;157:107344.
5. Zhu B, Hsieh CW, Zhang Y. Incorporating Spatial Statistics into Examin‑
ing Equity in Health Workforce Distribution: An Empirical Analysis in
the Chinese Context. Int J Environ Res Public Health. 2018;15(7).
6. Song X, Wei Y, Deng W, Zhang S, Zhou P, Liu Y, et al. Spatio‑Temporal Dis‑
tribution, Spillover Effects and Influences of China’s Two Levels of Public
Healthcare Resources. Int J Environment Res Public Health. 2019;16(4).
7. Vogl T, Lleras‑Muney A, Cutler DM. Socioeconomic Status and Health:
Dimensions and Mechanisms. National Bureau of Economic Research; 2008.
8. Sala‑i‑Martin X. On the health poverty trap. Health and Economic Growth:
Findings and Policy Implications. 2005:95‑114.
9. Zhang T, Xu Y, Ren J, Sun L, Liu C. Inequality in the distribution of health
resources and health services in China: hospitals versus primary care
institutions. Int J Equity Health. 2017;16(1):42.
10. Yip W, Fu H, Chen AT, Zhai T, Jian W, Xu R, et al. 10 years of health‑care
reform in China: progress and gaps in Universal Health Coverage. Lancet
(London, England). 2019;394(10204):1192–204.
11. Friel S, Marmot MG. Action on the social determinants of health and
health inequities goes global. Ann Review Public Health. 2011;32:225–36.
12. Yip WC, Hsiao WC, Chen W, Hu S, Ma J, Maynard A. Early appraisal of
China’s huge and complex health‑care reforms. Lancet (London, Eng‑
land). 2012;379(9818):833–42.
13. “Healthy China 2030” Planning Outline: General Office of the State
Council,PRC; 2016 [Available from: http:// www. gov. cn/ zheng ce/ 2016‑ 10/
25/ conte nt_ 51241 74. htm.
14. Tan X, Zhang Y, Shao H. Healthy China 2030, a breakthrough for improv‑
ing health. Global Health Promotion. 2019;26(4):96–9.
15. 2018 Medical Security Development Statistical Bulletin: China Healthcare
Security Administration; 2019 [Available from: http:// www. nhsa. gov. cn/
art/ 2019/2/ 28/ art_7_ 942. html.
16. Liu Y, Jiang Y, Tang S, Qiu J, Zhong X, Wang Y. Analysis of the equity of
emergency medical services: a cross‑sectional survey in Chongqing city.
Int J Equity Health. 2015;14:150.
17. Yan K, Jiang Y, Qiu J, Zhong X, Wang Y, Deng J, et al. The equity of China’s
emergency medical services from 2010‑2014. Int J Equity Health.
2017;16(1):10.
18. Zhang Y, Wang Q, Jiang T, Wang J. Equity and efficiency of primary health care
resource allocation in mainland China. Int J Equity Health. 2018;17(1):140.
19. Pan J, Shallcross D. Geographic distribution of hospital beds through‑
out China: a county‑level econometric analysis. Int J Equity Health.
2016;15(1):179.
20. 2018 Statistical Communique on the Development of China’s Health
Undertakings: National Health Commission; 2019 [Available from: http://
www. gov. cn/ guoqi ng/ 2020‑ 04/ 29/ conte nt_ 55075 28. htm.
21. Juran S, Broer PN, Klug SJ, Snow RC, Okiro EA, Ouma PO, et al. Geospatial
mapping of access to timely essential surgery in sub‑Saharan Africa. BMJ
Global Health. 2018;3(4):e000875.
22. Xu Y, Fu C, Onega T, Shi X, Wang F. Disparities in Geographic Accessibility
of National Cancer Institute Cancer Centers in the United States. J Med
Syst. 2017;41(12):203‑.
23. Esquivel MM, Uribe‑Leitz T, Makasa E, Lishimpi K, Mwaba P, Bowman K,
et al. Mapping Disparities in Access to Safe, Timely, and Essential Surgical
Care in Zambia. JAMA Surg. 2016;151(11):1064–9.
24. Miles RC, Onega T, Lee CI. Addressing Potential Health Disparities in
the Adoption of Advanced Breast Imaging Technologies. Acad Radiol.
2018;25(5):547–51.
25. Parmar D, Banerjee A. How do supply‑ and demand‑side interventions
influence equity in healthcare utilisation? Evidence from maternal health‑
care in Senegal. Soc Sci Med. 2019;241:112582.
26. Jing Q, Tang Q, Sun M, Li X, Chen G, Lu J. Regional Disparities of Rehabili‑
tation Resources for Persons with Disabilities in China: Data from 2014 to
2019. Int J Environment Res Public Health. 2020;17(19).
27. Mao Y, Zhang N, Zhu B, Liu J, He R. A descriptive analysis of the Spatio‑
temporal distribution of intestinal infectious diseases in China. BMC Infect
Dis. 2019;19(1):766.
28. He R, Zhu B, Liu J, Zhang N, Zhang WH, Mao Y. Women’s cancers in
China: a spatio‑temporal epidemiology analysis. BMC Women’s Health.
2021;21(1):116.
29. Wang J, Wang X, Li H, Yang L, Li Y, Kong C. Spatial distribution and deter‑
minants of health loss from Kashin‑Beck disease in Bin County, Shaanxi
Province, China. BMC Public Health. 2021;21(1):387.
30. Tao Y, Henry K, Zou Q, Zhong X. Methods for measuring horizontal equity
in health resource allocation: a comparative study. Health Econ Rev.
2014;4(1):10.
31. Zhang Q, Wang Y. Using concentration index to study changes in socio‑
economic inequality of overweight among US adolescents between
1971 and 2002. Int J Epidemiol. 2007;36(4):916–25.
32. Siegel M, Allanson P. Longitudinal analysis of income‑related health
inequalities: methods, challenges and applications. Expert Review Phar‑
macoeconomics Outcomes Research. 2016;16(1):41–9.
33. Qin X, Hsieh C‑R. Economic growth and the geographic maldistribu‑
tion of health care resources: Evidence from China, 1949‑2010. China
Economic Review. 2014;31:228–46.
34. Song C, Wang Y, Yang X, Yang Y, Tang Z, Wang X, et al. Spatial and Tempo‑
ral Impacts of Socioeconomic and Environmental Factors on Healthcare
Resources: A County‑Level Bayesian Local Spatiotemporal Regression
Modeling Study of Hospital Beds in Southwest China. Int J Environment
Res Public Health. 2020;17(16).
35. Li J, Chen X, Han X, Zhang G. Spatiotemporal matching between medical
resources and population ageing in China from 2008 to 2017. BMC Public
Health. 2020;20(1):845.
36. Wang JF, Li XH, Christakos G, Liao YL, Zhang T, Gu X, et al. Geographical
Detectors‑Based Health Risk Assessment and its Application in the Neural
Tube Defects Study of the Heshun Region, China. Int J Geographical
Information Sci. 2010;24(1):107–27.
37. Wang J‑F, Zhang T‑L, Fu B‑J. A measure of spatial stratified heterogeneity.
Ecological Indicators. 2016;67:250–6.
38. Páez A, Wheeler DC. Geographically Weighted Regression. In: Kitchin
R, Thrift N, editors. International Encyclopedia of Human Geography.
Oxford: Elsevier; 2009. p. 407–14.
39. Gittleman JL, Kot M. Adaptation: Statistics and a Null Model for Estimating
Phylogenetic Effects. Systematic Biology. 1990;39(3):227–41.
40. Ding J, Hu X, Zhang X, Shang L, Yu M, Chen H. Equity and efficiency
of medical service systems at the provincial level of China’s main‑
land: a comparative study from 2009 to 2014. BMC Public Health.
2018;18(1):214.
41. Global Health Expenditure Database [Internet]. World Health Organiza‑
tion. 2019 [cited 12 Dec 2020]. Available from: https:// apps. who. int/ nha/
datab ase/ Home/ Index/ en.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 13 of 13
Chenetal. BMC Public Health (2021) 21:2272
•
fast, convenient online submission
•
thorough peer review by experienced researchers in your field
•
rapid publication on acceptance
•
support for research data, including large and complex data types
•
gold Open Access which fosters wider collaboration and increased citations
maximum visibility for your research: over 100M website views per year
•
At BMC, research is always in progress.
Learn more biomedcentral.com/submissions
Ready to submit your research
Ready to submit your research
? Choose BMC and benefit from:
? Choose BMC and benefit from:
42. Notice on the work related to new rural cooperative medical care in
2011: Ministry of Health; 2011 [Available from: http:// www. nhc. gov. cn/
jws/ s3581 sg/ 201104/ 549c0 57f82 21479 88cbc 72ce7 07508 42. shtml.
43. Guidance of the General Office of the State Council on further deepening
the basic medical insurance payment reform: General Office of the State
Council; 2017 [Available from: http:// www. gov. cn/ zheng ce/ conte nt/
2017‑ 06/ 28/ conte nt_ 52063 15. htm.
44. China Statistical Yearbook 2019: National Bureau of Statistics of China;
[Available from: http:// www. stats. gov. cn/ tjsj/ ndsj/ 2019/ index ch. htm.
45. Liu H, Zhu H, Wang J, Qi X, Zhao M, Shan L, et al. Catastrophic health
expenditure incidence and its equity in China: a study on the initial
implementation of the medical insurance integration system. BMC Public
Health. 2019;19(1):1761.
46. Li D, Zhou Z, Si Y, Xu Y, Shen C, Wang Y, et al. Unequal distribution of
health human resource in mainland China: what are the determinants
from a comprehensive perspective? Int J Equity Health. 2018;17(1):29.
47. Witthayapipopsakul W, Cetthakrikul N, Suphanchaimat R, Noree T,
Sawaengdee K. Equity of health workforce distribution in Thailand:
an implication of concentration index. Risk Manag Healthc Policy.
2019;12:13–22.
48. A plan for the construction of primary health care teams focusing on
general practitioners: Ministry of Health; 2010 [Available from: http://
www. gov. cn/ gzdt/ 2010‑ 04/ 01/ conte nt_ 15713 24. htm.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub‑
lished maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com