Potential and realized access to healthcare services in Wuhan Metropolitan Area,China

Abstract

Healthcare accessibility is a crucial livelihood issue directly related to people’s health. Given that existing studies have often focused on residents’ potential geographic access in a single city, this study examines the access to healthcare services in Wuhan Metropolitan Area in both potential and realized dimensions. We pay particular attention to the quantitative measurement of the actual fit between healthcare services being offered and utilized. Overall, we found satisfactory potential access to healthcare facilities within reach of residential spaces in Wuhan Metropolitan Area, yet it is uneven across the nine cities and different areas. By measuring realized access through the coupling coordination degree of healthcare provision and utilization, our findings showed that except for Wuhan, all the other cities suffer from various levels of incoordination and lack resilience to public health emergencies. Taken together, this requires the regional government to formulate a more coordinated and comprehensive plan for resource allocation and to promote resource sharing among cities. Targeted policies for certain cities are also needed, to improve residents’ healthcare accessibility.

Keywords

Accessibility healthcare services Wuhan Metropolitan Area China coupling coordination degree

Introduction

Growing population density, rapid urbanization, and an irreversibly aging population have been posing challenges in providing essential services, especially healthcare, to all segments of the population (Giles-Corti et al., 2016). The unprecedented devastation brought about by the global outbreak of the COVID-19 pandemic has called researchers and policymakers to rethink approaches to this concerning problem, since healthcare accessibility is critical in assisting residents and communities to recover from such disasters (Haldane et al., 2021). For example, the COVID-19 outbreaks in China during early 2020 led to a severe shortage of hospital beds (Zhuang et al., 2021), which adversely affected the efficiency of healthcare system. The World Health Organization (WHO, 2020) reported that there were globally 651.92 million confirmed cases of COVID-19 as of the end of 2022, including 6.66 million deaths. Wuhan, with 50,008 COVID-19 cases as of April 8, 2020 (Xinhua News Agency, 2020), when the city government announced the end of the first lockdown, was the first city to be faced with such a public health emergency. As the core city of Wuhan Metropolitan Area (WMA), Wuhan attracted 3.95 million migrants in 2020, of whom 68.29% originated from other cities in Hubei province; thus, migrants accounted for over 32% of the total permanent residents (Wuhan Municipal Bureau of Statistics, 2021a, 2021b). This undoubtedly facilitated the rapid spread of COVID-19 from Wuhan to the other cities in WMA, during the 2020 Spring Festival traffic rush (Chen et al., 2020a; Wan and Wan, 2022). In this context, it is crucial to understand the extent to which healthcare services are available to the public in the whole Wuhan Metropolitan Area.

Researchers studying accessibility to urban amenities have aimed to understand the geographies of convenience or easiness that enable different segments of the population to use public services (McLafferty, 1982; Rahman et al., 2022; Sherman et al., 2005; Talen and Anselin, 1998). Scholars have explored the application of various data and methods to measure the relationship between the service delivery and individuals’ access to these services. Geospatial data are widely used in the measurement of geographic accessibility, regulated by space or distance, that allows individuals to easily reach their destination (Ashik et al., 2020; Dejen et al., 2019; Rahman et al., 2022). A diverse range of methods, including gravity approaches and cumulative opportunity measures, have been employed to evaluate spatial accessibility (Rahman et al., 2022). Aforementioned studies largely hinge on the assumption that services falling within residents’ catchment area are fully available to them (Wang and Luo, 2005), and therefore examine potential access (Khan, 1992). Statistical data on the quality and use of public services, on the other hand, offer the opportunity to examine residents’ actual choice and assess the relationship between the provision and utilization of public services (Chen et al., 2020b; Haynes et al., 2003). Researchers have used various techniques, such as the Gini coefficient and Robin Hood Index, to measure the degrees of equity and coordination of healthcare service allocation (Ren et al., 2021; Shadmi et al., 2014). While both potential and actual fit between healthcare services being offered and utilized is important for analyzing and improving residents’ access to essential services, few studies have combined the two dimensions.

Moreover, China’s radical urban transformation, at an unprecedented scale and speed, has led to the emergence of urban agglomerations (Zeng et al., 2015), which have fundamentally changed the organization of socioeconomic activities. However, the unevenness of economic development and resource allocation within the megalopolis or urban agglomeration has been a concern, especially regarding disparities between urban and rural areas (He and Zhang, 2022), and between large and small cities. In response to this spatial inequality, the State Council of China issued National New Urbanization Plan (2014–2020) in 2014 to “optimize the urbanization patterns” and promote more balanced development (Xinhua News Agency, 2014). Despite increasingly close collaborative networks in economic development (Fukuda, 2020), technology (Ma, 2022), and the ecological environment (Chen et al., 2022) in urban agglomerations, there have been few explorations of the intercity sharing of healthcare services in China. Even a decade after the Hubei province government issued Implementation Opinions on the Integrated Development of Healthcare Services in Wuhan Metropolitan Area in 2010, the number of healthcare institutions and beds in Wuhan in 2020 accounted for 54.2% and 42.3% of the entire metropolitan area, respectively (Hubei Provincial Statistics Bureau, 2021). The distribution of medical resources in WMA shows a high degree of polarization, which undermines the healthcare delivery system’s overall capacity. Therefore, the equitable allocation of healthcare resources at the regional level deserves greater consideration. However, only a handful of studies have studied the accessibility of healthcare services at the metropolitan area scale (Shen and Sun, 2022), and even fewer have paid attention to the central region of China.

Hence, there is currently a research gap in the evaluation of access to healthcare service in urban agglomerations. Accordingly, this paper aims to contribute to the existing literature in the following aspects: first, we measure the residents’ access to healthcare service in WMA in both potential and realized dimensions. Our particular focus on the quantitative measurement of the actual fit between the provision and utilization of healthcare services can offer important policy implications. Second, we establish a coupling coordination model to evaluate residents’ realized access to medical resources.

The remainder of this paper is organized as follows. The next section reviews relevant literature on the conceptualization and measurements of healthcare accessibility. The section after that describes the study area, as well as the data and methods utilized in this paper, followed by a section that presents major findings. After a discussions of the results, the paper concludes with the implications for regional healthcare resource allocation.

Literature review

Accessibility, as an important indicator of the efficiency, equity, and quality of public service delivery, has received extensive interest from academics and policymakers since the 1970s (Dalvi and Martin, 1976; Ingram, 1971). Researchers from a wide range of disciplines, including planning, public policy, and geography, have offered theoretically and operationally sound definitions of accessibility. In studies on access to healthcare services, “accessibility” has been understood as an attribute of an institution or service that is available whenever and wherever populations need it (Aday and Andersen, 1974; Donabedian, 1966). It describes characteristics—geographic and socio-organizational—that facilitate or hinder people’s efforts to utilize healthcare services (Donabedian, 1973). Based on this dichotomy between geographic and social-organizational aspects, two strands of studies have emerged and persisted in the literature on measuring health service accessibility.

Several researchers have explored the role of unevenly distributed physical or environmental features in affecting individuals’ ease of accessing the system (Guagliardo, 2004; Joseph and Phillips, 1984; Luo and Wang, 2003; Shannon et al., 1975). Many geographic studies use the “friction of space,” a function of the time and physical distance, to conceptualize and measure the spatial dimension of healthcare accessibility (Aday and Andersen, 1974). The progress in Geographic Information Systems (GIS) and spatially referenced data have led to a wide array of location-based accessibility measures. The most intuitive method is the use of the contour measure, also known as the isochronic measure, or cumulative opportunities, to count the number of reachable health services within a given catchment area, as defined by travel time, distance, or cost (Geurs and Van Wee, 2004). This measure has advantages in terms of its computability and interpretability, but suffers from poor theoretical grounding, since it neglects the effects of distance decay (Joseph and Bantock, 1982), and does not take into account individuals’ preferences. Hence, gravity-based measures have been introduced to more validly define catchment areas by discounting the accessibility with increasing distance or travel impedance (Guagliardo, 2004). The popular two-step floating catchment area also allows researchers to consider the interaction between supply and demand for health services (McGrail, 2012; Wang, 2012, 2018; Wang and Luo, 2005). Additionally, researchers have managed to incorporate additional travel information, including the street network, speed limits, travel time, and traffic lights, into these measures to evaluate accessibility (Kanuganti et al., 2016; Martin et al., 2002; Wang and Luo, 2005). For instance, by utilizing online map APIs to improve the travel time estimation, Tao et al. (2018) used the multi-modal two-step floating catchment area (2SFCA) method to identify regions with a shortage of healthcare services in Shenzhen, China.

In studying the multi-dimensional nature of healthcare accessibility, other researchers focus on characteristics of the population or of the structure, which render medical resources more or less easy to use (Andersen, 1995; Hawthorne and Kwan, 2013). First, it is of critical importance to consider individuals’ multiple axes of differences when assessing accessibility. These individual determinants (Andersen and Davidson, 2014; Andersen and Newman, 1973) not only include characteristics that reflect the healthcare needs of different groups, such as age and gender; they also incorporate factors describing resources specific to them, such as income and insurance coverage, as well as factors in their probability of seeking care, such as their attitudes toward and knowledge about healthcare (Aday and Andersen, 1974). Second, the structural features of the healthcare delivery system, including resources and organization, also determine whether entry can be efficiently gained. Measures of the health service delivery system focus on (a) the volume and distribution of healthcare resources, which includes medical manpower and facilities; and (b) the manner in which these resources are coordinated and controlled to offer health services (Andersen and Davidson, 2014; Coleman, 1975). These measures can be derived from indicators of health service capacity, such as the number of physicians, hospital beds, or institutions.

Nevertheless, while the above measurements investigate accessibility from diverse entry points, ranging from costs (e.g., physical or time distance) to psychological variables or health knowledge, most studies focus solely on the provision of medical service, and therefore examine potential access (Guagliardo, 2004; Khan, 1992). In other words, while potential access indicates the availability of healthcare services to residents, it does not signify an “automatic utilization” of these available services (Khan, 1992: 275). Only when residents actually utilize the service offered by the healthcare system is the realized access achieved, which is also referred to as a fit between the system and the patient (Penchansky and Thomas, 1981). Therefore, despite its validity in revealing residents’ ability to get care, research on potential access could be problematic in its neglect of the gap between the potential access offered by the healthcare system and the realized access by the residents (Joseph and Phillips, 1984; Lin et al., 2005). Outcome indicators of individuals’ actual entry into the system, such as utilization rates or satisfaction scores, are thus needed to permit “external validation” of the effectiveness of the healthcare system (Aday and Andersen, 1974; Khan, 1992). In this sense, the coupling coordination degree model—a concept that quantifies the degree of harmony between subsystems within the main system (Li et al., 2012, 2022)—is applicable to analyze the fit between health services being offered and utilized.

In addition, existing studies typically focus on the healthcare accessibility in a single city (Huang et al., 2019; Luo and Wang, 2003; Yang et al., 2006), but fail to take into account the emergence of urban agglomerations consisting of a group of cities, where healthcare resources complement or compete with each other (Guida and Carpentieri, 2021) and bring about increased intercity patient mobility (Andritsos and Tang, 2014; Ding et al., 2023). More importantly, achieving equity in healthcare accessibility is a challenge for decision-makers at every territorial scale (Guida and Carpentieri, 2021; Taleai et al., 2014; Weiss et al., 2018). The development of public health in China, for instance, has entered a new stage of striving for an equitable healthcare system, but the inequality of healthcare resource allocation remains prominent (Wang et al., 2019; Yin et al., 2018). Putting these two points together, regional urban agglomeration is an appropriate scale for studying the regional synergy of healthcare resources. However, previous studies have paid relatively limited attention to the regional heterogeneity of healthcare accessibility. Ghorbanzadeh et al. (2021) identified the high and low healthcare accessibility areas in the entire state of Florida, showing that northwest and southern Florida suffer from insufficient health services, while central Florida has the highest level of accessibility. Li and Wang (2022) evaluated the healthcare accessibility to high-tier medical institutions in the Changsha-Zhuzhou-Xiangtan urban agglomeration, China.

Recognizing that healthcare accessibility is a multi-dimensional and multi-scale concept, we seek to evaluate residents’ access to healthcare services in Wuhan Metropolitan Area in both potential and realized dimensions. For this purpose, this study firstly uses spatial analysis methods to measure residential spaces’ healthcare accessibility in each city and the whole Wuhan Metropolitan Area. Secondly, we employ the entropy-weighted TOPSIS model and coupling coordination model to measure the realized access to healthcare services. Our findings can offer useful insights for Wuhan Metropolitan Area and other urban agglomerations, to improve regional health service delivery systems. The methodology and results will be discussed in detail in the following sections.

Study area and methodology

Study area

The study focuses on Wuhan Metropolitan Area, Hubei Province, China. Comprising nine cities¹ (Wuhan, Huangshi, Ezhou, Xiaogan, Xianning, Huanggang, Xiantao, Qianjiang, and Tianmen), it is located on the middle reaches of the Yangtze River (see Figure 1). As the largest economically developing city cluster in Central China (Wen et al., 2021), WMA covers only 31.2% of the area of Hubei province, yet it concentrates a population exceeding 31.98 million and contributes to 60.38% of the province’s economic output (Hubei Provincial Statistics Bureau, 2021). According to Implementation Opinions on the Integrated Development of Wuhan Metropolitan Area issued in 2022, WMA strives to improve the sharing of public services, in which the optimization of healthcare resource allocation is of vital importance. In addition, Wuhan was the epicenter of the start of the COVID-19 pandemic in early 2020, followed by outbreaks in WMA. The overburdened healthcare system not only led to decreased quality of care, but also alerted the government to the urgent need for a more resilient healthcare system in WMA (Li et al., 2020).

Figure 1.

The location of Wuhan Metropolitan Area (WMA): (a) its location in China and (b) its location in Hubei Province.

Data

This study employed various data sources, including geographic and statistical data of healthcare services in Wuhan Metropolitan Area (see Table 1). Firstly, geographic data of healthcare services and residential spaces were scraped from Point-of-Interest (POI) datasets of the Gaode Map (AMap) Geocoding API in 2022. After de-duplicating, screening, and matching the datasets, we obtained POI footprint data of 32,891 healthcare facilities and 26,029 residential spaces. Among them, healthcare facilities include pharmacies, clinics, disease control and prevention centers, emergency rooms, specialized hospitals, and general hospitals; residential spaces include residential communities, residential–commercial hybrid buildings, dormitories, and industrial parks in urban areas, as well as villages and settlements in rural areas. Secondly, we used statistical data of healthcare services to measure the coupling coordination degree of medical service provision and utilization in each city in WMA during 2019–2020. The Healthcare Resource Density Index (HRDI) is employed to measure the healthcare resource provided by each city, including institutions, hospital beds, and healthcare technicians (Zhang et al., 2018). The health service utilization is measured according to the function of healthcare institutions, and therefore, considers the health consultations and inpatient operations, as shown in Table 1.

Table 1.

Data sources.

Dimensions	Data	Source
Potential access	POI of healthcare facilities and residential spaces	A
Realized access	The number of healthcare facilities in each city	B
	The number of tertiary hospitals in each city	B
	The number of healthcare technicians in each city	B
	The number of hospital beds in each city	B
	The average annual number of health consultations per capita in each city	B
	The annual hospital bed utilization rate (%) in each city	C
	The average annual number of inpatient days per capita in each city	C
	The average number of health consultations in charge per technician per day in each city	C
	The average number of inpatient days in charge per technician per day in each city	C

Data sources: A is from the Gaode Map (AMap) Geocoding API; B is from the 2020, 2021 Statistical Yearbook in each city; C is from the Health Commission of Hubei Province (http://wjw.hubei.gov.cn/).

Methodology

GIS spatial analysis

Four important GIS spatial analysis tools were employed in this paper to measure the residents’ potential access to healthcare services in Wuhan Metropolitan Area: 1) standard deviational ellipse; 2) average nearest neighbor; 3) kernel density; and 4) buffer analysis.

First, standard deviational ellipse is a common way to measure the spatial distribution of a set of points by calculating the standard deviation of the x-coordinates and y-coordinates from the mean center. The standard deviational ellipse is given as:

C = (\begin{matrix} v a r (x) & c o v (x, y) \\ c o v (y, x) & v a r (y) \end{matrix}) = \frac{1}{n} (\begin{matrix} \sum_{i = 1}^{n} {(x_{i} - \bar{x})}^{2} & \sum_{i = 1}^{n} (x_{i} - \bar{x}) (y_{i} - \bar{y}) \\ \sum_{i = 1}^{n} (x_{i} - \bar{x}) (y_{i} - \bar{y}) & \sum_{i = 1}^{n} {(y_{i} - \bar{y})}^{2} \end{matrix})

(1)

where $x_{i}$ and $y_{i}$ are the coordinates for feature $i,$ $\bar{x}$ and $\bar{y}$ represent the coordinates for mean center of the features, and n is the total number of features.

Second, the average nearest neighbor is a measure of the distance between each feature’s centroid and its nearest neighbor’s centroid. All these nearest neighbor distances are averaged and compared with a hypothetical random distribution (expected distribution) to reveal whether the features are clustered or dispersed. The average nearest neighbor ratio is calculated as:

A N N = \frac{{\bar{D}}_{O}}{{\bar{D}}_{E}}

(2)

where

{\bar{D}}_{O} = \frac{\sum_{i = 1}^{n} d_{i}}{n}

(3)

{\bar{D}}_{E} = \frac{0.5}{\sqrt{n / A}}

(4)

In the above equations, ${\bar{D}}_{O}$ denotes the observed mean distance between each feature and its nearest neighbor, and ${\bar{D}}_{E}$ denotes the expected distance of the features given in a random pattern; $d_{i}$ equals the distance between feature i and its nearest neighboring feature, n is the total number of features, and A is the area of a minimum enclosing rectangle around all features. If the index $A N N$ is less than one, the spatial pattern of features exhibits clustering. If the index is greater than one, the spatial pattern is considered dispersed.

Third, the kernel density tool has been applied to better assess healthcare service accessibility (Spencer and Angeles, 2007; Yang et al., 2006). It calculates the density of point features in a neighborhood around those features, and creates a smoothly curved surface fitting over each point. The surface value is highest at the location of the point and diminishes with increasing distance from the point, reaching zero at the search radius distance from the point. Referring to official guidelines for healthcare service planning,² this study chose five-minute walking distance (300 meters) and 15-minute walking distance (1000 meters) as the search radius distances for primary-level healthcare facilities (including clinics and community hospitals), and higher-level healthcare facilities (including emergency rooms, general hospitals, and specialty hospitals), respectively.

Accordingly, we created buffer polygons around healthcare facilities and residential spaces. We counted 1) the residential communities within the healthcare facilities’ catchment area and 2) the primary-level and higher-level healthcare facilities within the residential spaces’ catchment area. Furthermore, we created an overlay map of buffer analysis showing the areas with excess, sufficient, and insufficient healthcare resources in WMA: 1) if there is no residential space within the healthcare facilities’ catchment area, the area was defined as with excess healthcare resources; 2) if there are either primary-level or higher-level facilities within the residential spaces’ catchment area, the area was defined as with sufficient healthcare resources; and 3) if there is no healthcare facility within the residential spaces’ catchment area, the area was defined as with insufficient healthcare resources.

Entropy-weighted TOPSIS method and the coupling coordination degree model

Penchansky and Thomas (1981) emphasized that accessibility is the fit between a patient and the healthcare system. The entropy weighted TOPSIS method can objectively evaluate the level of healthcare resource provision and utilization (Sun et al., 2017), based on which the coupling coordination model allows us to measure the fit (Li et al., 2022); that is, residents’ realized access to healthcare services in WMA. Previous studies have applied the coupling coordination model to assess the supply and demand for urban green growth (Li et al., 2022), early care service (Zhang et al., 2022), and metro systems (Zhu et al., 2021). This study is among the first ones that apply entropy-weighted TOPSIS method and coupling coordination model to the evaluation of residents’ realized access to healthcare services. The evaluation process can be decomposed into the seven steps below:

(1) Constructing the initial matrix and standardizing it

Assuming there are m cities and n indicators in the index system, the value of indicator j in city i is $a_{i j} (i = 1, 2, \dots, m, a n d j = 1, 2, \dots, n) .$ The initial matrix is:

A = {(a_{i j})}_{m \times n} = [\begin{matrix} a_{11} & \dots & a_{1 n} \\ ⋮ & ⋱ & ⋮ \\ a_{m 1} & \dots & a_{m n} \end{matrix}]

(5)

As all indicators (see Table 6) are positive in this paper, we used formula (6) to standardize the initial values $a_{i j}$ to a notionally common scale and the standardized matrix is $B = {(b_{i j})}_{m \times n} :$

b_{i j} = \frac{a_{i j} - m i n {a_{1 j}, \dots, a_{m j}}}{m a x {a_{1 j}, \dots, a_{m j}} - m i n {a_{1 j}, \dots, a_{m j}}}

(6)

(2) Determining the indicator’s weight

In information theory, the entropy is used to quantify the average level of information or uncertainty (Gao et al., 2020; Huang, 2008) inherent to the indicator’s possible outcomes. It can be calculated as:

\begin{array}{l} e_{j} = - \frac{1}{l n m} \sum_{i = 1}^{m} (\frac{b_{i j}}{\sum_{i = 1}^{m} b_{i j}} l n \frac{b_{i j}}{\sum_{i = 1}^{m} b_{i j}}) \end{array}

(7)

A larger $e_{j}$ indicates a greater differentiation degree of indicator $j,$ which means the indicator carries more information (Zhu et al., 2020). Hence, higher weight should be given to the indicator and the weight can be calculated as:

w_{j} = \frac{1 - e_{j}}{\sum_{j = 1}^{n} (1 - e_{j})}

(8)

(3) Constructing the weighted matrix

The weighted value for indicator j in city i can be calculated using formula (9):

v_{i j} = w_{j} \times b_{i j}

(9)

The weighted matrix can accordingly be represented as:

V = {(v_{i j})}_{m \times n} = [\begin{matrix} w_{1} b_{11} & \dots & w_{n} b_{1 n} \\ ⋮ & ⋱ & ⋮ \\ w_{1} b_{m 1} & \dots & w_{n} b_{m n} \end{matrix}]

(10)

(4) Determining positive and negative ideal solutions

V^{+} = {v_{1}^{+}, v_{2}^{+}, \dots, v_{n}^{+}} = {(m a x v_{i j} | j \in J) | i = 1, 2, \dots, m}

(11)

V^{-} = {v_{1}^{-}, v_{2}^{-}, \dots, v_{n}^{-}} = {(m i n v_{i j} | j \in J) | i = 1, 2, \dots, m}

(12)

Where $V^{+}$ denotes the positive ideal solutions, and $V^{-}$ denotes the negative ideal solutions.

(5) Calculating relative distance

This study used Euclidian distance in formulas (13) and (14) to calculate the distance between the weighted value of indicator j in city i to the indicator’s ideal solutions:

S_{i}^{+} = \sqrt{\sum_{j = 1}^{n} {(v_{j}^{+} - v_{i j})}^{2}} (i = 1, \dots, m)

(13)

S_{i}^{-} = \sqrt{\sum_{j = 1}^{n} {(v_{j}^{-} - v_{i j})}^{2}} (i = 1, \dots, m)

(14)

where $S_{j}^{+}$ ( $S_{j}^{-}$ ) denotes the distance between $v_{i j}$ and positive (negative) ideal solution.

(6) Calculating the degree of approximation

E_{i} = \frac{S_{i}^{-}}{S_{i}^{+} + S_{i}^{-}} (0 \leq E_{i} \leq 1, i = 1, 2, \dots, m)

(15)

where $E_{i}$ denotes the degree of approximation for city $i .$ The larger $E_{i}$ is, the better the healthcare resources in city i are allocated or utilized.

(7) Calculating the coupling coordination degree

Using Equations (5)–(15) and indicators in Table 6, we calculated the healthcare resource provision index $X_{i}$ and the healthcare service utilization index $Y_{i}$ for each city. Finally, equations (16) allowed us to calculate the coupling coordination of healthcare resource provision and utilization in city $i :$

{\begin{matrix} D_{i} = \sqrt{C_{i} \times T_{i}}, D_{i} \in (0, 1] \\ w h e r e C_{i} = \sqrt{(X_{i} \times Y_{i}) \times {(\frac{X_{i} + Y_{i}}{2})}^{- 2}}, \\ a n d T_{i} = α X_{i} + β Y_{i} \end{matrix}

(16)

where $C_{i}$ is the coupling degree, i.e., the degree of interdependence between $X_{i}$ and $Y_{i}$ ; and $T_{i}$ reflects a comprehensive evaluation of healthcare system in city i, where α and β are the contribution coefficients. Considering the equal importance of service provision and utilization to the healthcare system, the values of α and β are equivalent, i.e., $α = β = 0.5$ . $D_{i}$ denotes the coupling coordination degree of healthcare resource provision and utilization in city i. By incorporating $C_{i}$ and $T_{i}$ , the magnitude $D_{i}$ not only reflects how closely connected $X_{i}$ and $Y_{i}$ are, but also the relative capacity and utilization of the healthcare services. A larger $D_{i}$ points toward a relatively higher level of coordination between the provision and utilization of healthcare services in city i, compared with other cities in the WMA but the type of the coupling coordination degree is also important as it reveals the origin of possible incoordination. Therefore, based on the obtained coupling coordination degree $D_{i}$ and healthcare service provision and utilization index ( $X_{i}$ and $Y_{i}),$ we adopted the standards of the coupling coordination degree (Shi et al., 2020) to classify the results into six levels and three types, as shown in Table 2.

Table 2.

The standard of the coupling coordination degree.

D value	Level	Index	Type
0.8<D_i⩽1	Quality coordination(V1)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag
0.6<D_i⩽0.8	Intermediate coordination(V2)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag
0.5<D_i⩽0.6	Primary coordination(V3)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag
0.4<D_i⩽0.5	Basic incoordination(V4)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag
0.2<D_i⩽0.4	Intermediate incoordination(V5)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag
0<D_i⩽0.2	Extreme incoordination(V6)	X_i>Y_i	Utilization Lag
		X_i=Y_i	Utilization-Provision synchronized
		X_i<Y_i	Provision Lag

Findings

Residents’ potential access to healthcare services

First, standard deviational ellipse analysis results shown in Figure 2 reveal the distributional trend of healthcare facilities and residential spaces in Wuhan Metropolitan Area. The dispersion patterns of healthcare resources and residential spaces are consistent in each city and in WMA. More specifically, as shown in Figure 2a, the distribution direction of healthcare facilities and residential spaces in WMA shows an east–west trend, with Wuhan as the center. At the city level, medical and residential points are distributed in various directions in different cities (see Figure 2b). For example, the distribution of the two types of spaces in Xiantao and Qianjiang is oriented east–west, while the layouts of medical and residential spaces in Xiaogan, Huangshi, Ezhou, and Huanggang are in the northeast direction. Such differences in the spatial orientation can be attributed to the cities’ geometry, topography, and location. In addition, the differences in the size of the standard ellipses illustrated by Figure 2 show that healthcare facilities in WMA are more dispersed than residential spaces.

Figure 2.

Standard deviational ellipse analysis of healthcare facilities and residential spaces in WMA at (a) regional level analysis and (b) city level analysis.

Second, Table 3 displays the results of average nearest neighbor analysis of the healthcare facilities and residential spaces in WMA. The average nearest neighbor ratios are less than 1, indicating clustering patterns of healthcare resources and residential spaces. Moreover, healthcare facilities and residential spaces are clustered to different degrees among the nine: they are more clustered in Xianning and Xiaogan, while less so in Wuhan, Xiantao, and Ezhou. Also note that the spatial distribution of healthcare resources is typically less clustered than that of residential spaces; this finding is consistent with the results of the standard deviational ellipse.

Table 3.

Average nearest neighbor analysis results.

Area	Space	Nearest-neighbor ratio	Average nearest-neighbor distance	Average expected distance	Z Score
WMA	Healthcare Facilities	0.199869	192.4711	962.9871	-277.607
WMA	Residential Places	0.248618	266.8610	1073.3780	-231.911
Wuhan	Healthcare Facilities	0.249773	129.2779	517.5816	-177.755
Wuhan	Residential Places	0.343236	168.9903	492.3451	-156.627
Huangshi	Healthcare Facilities	0.235644	213.7936	907.2727	-71.0968
Huangshi	Residential Places	0.335653	383.8809	1143.6843	-49.2726
Ezhou	Healthcare Facilities	0.322703	302.8860	938.5904	-42.1855
Ezhou	Residential Places	0.348339	312.5174	897.1653	-37.5867
Xiaogan	Healthcare Facilities	0.192056	219.3012	1141.8581	-98.2189
Xiaogan	Residential Places	0.261995	376.3510	1436.4829	-67.7838
Huanggang	Healthcare Facilities	0.200271	285.0734	1423.4417	-105.266
Huanggang	Residential Places	0.272175	540.2967	1985.1046	-70.5457
Xianning	Healthcare Facilities	0.186895	255.4685	1366.9111	-77.0732
Xianning	Residential Places	0.244351	411.6645	1684.7241	-56.5638
Xiantao	Healthcare Facilities	0.271494	250.5193	922.7422	-47.1592
Xiantao	Residential Places	0.276387	284.2300	1028.3781	-40.4545
Tianmen	Healthcare Facilities	0.220845	271.6749	1230.1609	-45.4076
Tianmen	Residential Places	0.36676	656.4314	1789.8104	-23.8932
Qianjiang	Healthcare Facilities	0.238124	240.5243	1010.0822	-41.9402
Qianjiang	Residential Places	0.307542	378.3041	1230.0878	-27.5019

Third, Figure 3 shows the kernel density analysis results for the healthcare facilities and residential spaces in WMA. From the perspective of the entire area, the highest density of both types of spaces can be observed in the central district of. Moreover, note that the density of high-level healthcare facilities such as general hospitals within a 15-minute range of the living area (Figure 3b) is higher than that of primary-level healthcare facilities, such as clinics, within five minutes of the living area (Figure 3a). For each city, the peak of the distribution density of both types of space is concentrated in the central area.

Figure 3.

Overlay map of kernel density analysis of healthcare facilities and residential spaces within (a) 5-minute walking distance and (b) 15-minute walking distance.

Fourth, Figures 4 and 5 present the results obtained via buffer analysis of healthcare facilities and residential spaces in WMA. Using the buffer distance of healthcare facilities as the horizontal coordinate, we counted the settlements within each buffer zone. As shown in Figure 4, the 1000-meter buffer of healthcare facilities in most cities can cover over 90% of the residential spaces. However, except for Qianjiang (100%) and Wuhan (80.91%), the 300-meter buffer zones of healthcare facilities cover around 70% of the residential spaces. Moreover, we overlay the buffer of healthcare facilities and residential spaces, and categorize the buffers into areas with excess, sufficient, and insufficient healthcare resources (Figure 5). It is noticeable that the distribution of medical and residential spaces is typically most coordinated in urban central areas, especially that of Wuhan. Nevertheless, it also reveals that healthcare facilities are inadequate in the urban–rural continuum, while they are in surplus in peripheral areas of some cities, such as Wuhan, Xiantao, Xianning, and Huanggang.

Figure 4.

Coverage of residential areas by healthcare resources at 300 m and 1000 m.

Figure 5.

Overlay map of buffer analysis.

Coupling coordination degree analysis of healthcare resources

Healthcare service provision and utilization

Tables 4 and 5 show the healthcare service provision and utilization in the Wuhan Metropolitan Area, respectively. Regarding the provision of healthcare services, Wuhan has concentrated most of the healthcare facilities, tertiary hospitals, healthcare technicians, and hospital beds. Although there is no significant change in healthcare resource provided by the city of Wuhan during 2019–2020, the density indices in other cities such as Xiaogan, Huanggang, and Xianning have increased. For healthcare service utilization, changes have been clear in these two years. The number of consultations per capita in Wuhan dropped from 8.13 in 2019 to 5.18 in 2020, and this trend is also reflected in other indicators such as hospital bed utilization rate. By contrast, as shown by the number of consultations, hospital utilization rate, and other indicators, healthcare resource utilization in Xiantao, Tianmen, and Qianjiang increased by varying degrees.

Table 4.

Healthcare service provision in WMA, 2019–2020.

Year	Healthcare service provision	City
Year	Healthcare service provision	Wuhan	Huangshi	Ezhou	Xiaogan	Huanggang	Xianning	Xiantao	Tianmen	Qianjiang
2019	Number of healthcare facilities	407	51	23	64	67	38	21	15	9
	Healthcare Facility Density Index	1.31	0.48	0.56	0.31	0.2	0.24	0.39	0.26	0.2
	Number of tertiary hospitals	27	8	4	6	6	4	2	2	2
	Tertiary Hospital Density Index	0.2	0.08	0.1	0.03	0.02	0.03	0.04	0.03	0.05
	Number of healthcare technicians	114200	19806	8276	26411	32366	17988	8062	5950	6453
	Healthcare Technician Density Index	368.43	185.95	201.24	126.17	97.34	114.1	149.87	104.04	146.66
	Number of hospital beds	96357	17767	5664	25630	38218	15760	7931	7064	4948
	Hospital Bed Density Index	310.87	166.81	137.73	122.44	114.94	99.97	147.44	123.52	112.45
2020	Number of healthcare facilities	362	51	23	69	66	40	28	16	12
	Healthcare Facility Density Index	1.11	0.48	0.55	0.35	0.21	0.25	0.52	0.29	0.28
	Number of tertiary hospitals	66	9	4	6	6	4	2	2	2
	Tertiary Hospital Density Index	0.19	0.08	0.1	0.03	0.02	0.02	0.04	0.04	0.05
	Number of healthcare technicians	115200	20181	8232	27500	36919	22937	8550	6452	6492
	Healthcare Technician Density Index	352.73	189.71	198.33	141.03	115.21	142.47	159.32	117.06	154.02
	Number of hospital beds	92800	17752	5847	26500	39532	16093	8332	7363	5036
	Hospital Bed Density Index	284.14	166.88	140.87	135.9	123.36	99.96	155.26	133.59	119.48

Table 5.

Healthcare service utilization in WMA, 2019–2020.

Year	Healthcare service utilization	City
Year	Healthcare service utilization	Wuhan	Huangshi	Ezhou	Xiaogan	Huanggang	Xianning	Xiantao	Tianmen	Qianjiang
2019	Average number of health consultations per capita	81286389	12765630	5090340	21285672	29455269	13991328	2243835	1986244	1346671
	Hospital bed utilization rate (%)	94.88	84.54	75.27	85.61	90.41	94.84	85.62	100.69	102.07
	Average number of inpatient days per capita	9.2	9	9.3	7.8	8.7	9	8.9	11.1	9.9
	Average number of health consultations in charge per technician per day	2.00	1.63	1.34	1.38	1.60	1.36	1.31	1.82	1.27
	Average number of inpatient days in charge per technician per day	6.16	6.61	7.26	7.61	6.60	6.35	10.07	9.92	8.50
2020	Average number of health consultations per capita	64502576	11040595	4037592	19031884	26095077	11705954	6828250	6794147	5232172
	Hospital bed utilization rate (%)	72.07	69.24	55.41	69.45	75.99	78.46	66.73	77.83	82.15
	Average number of inpatient days per capita	10.1	10.4	9.2	8.9	9.7	10.3	10.2	12.5	10.6
	Average number of health consultations in charge per technician per day	1.78	1.57	1.33	1.33	1.25	1.31	1.28	1.62	1.24
	Average number of inpatient days in charge per technician per day	8.5	7.94	9.23	8.80	9.56	7.90	3.38	3.26	2.24

Entropy-weighted TOPSIS evaluation results

The entropy method provided the indicator weights of healthcare service provision and utilization from 2019 to 2020, as shown in Table 6. In terms of healthcare service provision, the healthcare facility density index was the most influential indicator (29.79%), while the health technician density index contributed least (22.51%) in 2019. Nevertheless, in 2020, the tertiary hospital density index became the most important indicator, with its weight increasing to 28.82%. This reflects the change after the outbreak of COVID-19 in 2020, when tertiary hospital and healthcare technicians became more critical resources. In terms of healthcare resource utilization, the weights of the five indicators vary considerably in each year, changing drastically from 2019 to 2020. In both years, note that the average number of inpatient days per technician in charge stand out in supreme importance. Meanwhile, the weight of the number of health consultations per capita decreased from 52.85% in 2019 to 26.27% in 2020. In contrast, the weights of hospital bed utilization rate, average annual number of inpatient days, and average number of inpatient days in charge per technician per day increased significantly.

Table 6.

Weight of each indicator for healthcare service provision and utilization in WMA in 2019 and 2020.

Sub-System	Indicator	2019	2020
Healthcare service provision	Healthcare Facility Density Index	29.79%	24.38%
	Tertiary Hospital Density Index	24.35%	28.82%
	Healthcare Technician Density Index	22.51%	26.99%
	Hospital Bed Density Index	23.35%	19.81%
Healthcare service utilization	The average annual number of health consultations per capita in each city	52.85%	26.27%
	The annual hospital bed utilization rate (%) in each city	1.65%	12.12%
	The average annual number of inpatient days per capita in each city	1.70%	9.33%
	The average number of health consultations in charge per technician per day in each city	4.94%	17.28%
	The average number of inpatient days in charge per technician per day in each city	38.86%	34.99%

We further employed the entropy-weighted TOPSIS method to evaluate the healthcare service provision and utilization in each city of WMA during 2019–2020. Table 7 demonstrates wide variations in the provision and utilization of healthcare services across cities. The ranking of medical service provision in the metropolitan area remained unchanged from 2019 to 2020, with Wuhan consistently surpassing other cities by a wide margin in medical resources. However, it is also worth noting that except for Wuhan, the other eight cities experienced an increase in the healthcare service provision. In terms of healthcare service utilization, Wuhan was ranked first in 2019 but slipped to fourth in 2020. In contrast, Tianmen, Xiantao, and Qianjiang moved up from the last to the top three in medical resource utilization.

Table 7.

Evaluation of healthcare service provision and utilization in WMA in 2019 and 2020.

City	2019				2020
City	Provision X_i	Rank	Utilization Y_i	Rank	Provision X_i	Rank	Utilization Y_i	Rank
Wuhan	1.000	1	0.938	1	1.000	1	0.365	4
Huangshi	0.297	3	0.598	3	0.338	3	0.265	6
Ezhou	0.338	2	0.583	4	0.369	2	0.230	8
Xiaogan	0.093	6	0.522	6	0.129	6	0.360	5
Huanggang	0.032	9	0.572	5	0.048	9	0.233	7
Xianning	0.040	8	0.638	2	0.067	8	0.219	9
Xiantao	0.176	4	0.102	7	0.230	4	0.762	2
Tianmen	0.077	7	0.068	8	0.100	7	0.855	1
Qiangjiang	0.110	5	0.005	9	0.139	5	0.624	3

The dynamics of coupling coordination

Based on the entropy weighted TOPSIS evaluation results, we used the coupling coordination degree model to assess the interaction between healthcare service provision and utilization in Wuhan Metropolitan Area in 2019 and 2020. The results are classified into six levels and four types based on their coupling coordination degree and X_i–Y_i relationship, as illustrated by Tables 2 and 8 and Figure 6. Figure 6 visually reveals two outstanding insights. First, despite apparent variations across cities and periods, the results reveal an overall incoordination between healthcare service provision and utilization in WMA. While Wuhan, Huangshi, and Ezhou reached coordination in 2019, they also saw a decline in the coupling coordination degree in 2020. In fact, only Xiantao, Tianmen, and Qianjiang experienced a marked increase in coupling coordination degree from 2019 to 2020.

Figure 6.

Coupling coordination degree level (up) and type (down) of healthcare resource allocation and utilization in WMA in 2019 (left) and 2020 (right).

Table 8.

Levels and types of coupling coordination degree of healthcare service provision and utilization in WMA in 2019 and 2020.

City	2019			2020
City	D_i	Level	Type	D_i	Level	Type
Wuhan	0.984	Quality coordination(V1)	Utilization Lag	0.777	Intermediate coordination(V2)	Utilization Lag
Huangshi	0.649	Intermediate coordination(V2)	Provision Lag	0.547	Primary coordination(V3)	Utilization Lag
Ezhou	0.666	Intermediate coordination(V2)	Provision Lag	0.540	Primary coordination(V3)	Utilization Lag
Xiaogan	0.469	Basic incoordination(V4)	Provision Lag	0.464	Basic incoordination(V4)	Provision Lag
Huanggang	0.368	Intermediate incoordination(V5)	Provision Lag	0.325	Intermediate incoordination(V5)	Provision Lag
Xianning	0.400	Intermediate incoordination(V5)	Provision Lag	0.348	Intermediate incoordination(V5)	Provision Lag
Xiantao	0.366	Intermediate incoordination(V5)	Utilization Lag	0.647	Intermediate coordination(V2)	Provision Lag
Tianmen	0.269	Intermediate incoordination(V5)	Utilization Lag	0.541	Primary coordination(V3)	Provision Lag
Qiangjiang	0.153	Extreme incoordination(V6)	Utilization Lag	0.543	Primary coordination(V3)	Provision Lag

However, it is important to note that a high coupling coordination degree does not ensure a satisfactory realized access to healthcare services. For instance, while the level of coupling coordination degree in Tianmen reached primary coordination (V3) in 2020, it was due to a dramatic surge in the utilization of healthcare services in Tianmen. Note that it was ranked first in the utilization index but seventh in the provision index in 2020. This is why the second aspect, i.e., the type of coupling coordination degree model, is as important. Figure 6 presents various types of coupling coordination degree of healthcare service provision and utilization in WMA. It is noticeable that medical resources in Wuhan were consistently underutilized, while Xiaogan, Huanggang, and Xianning failed to provide adequate healthcare resources for their residents. In addition, Figure 6 captures changes in the coordination between healthcare service provision and utilization from 2019 to 2020. That is, the types of coupling coordination degree in Ezhou and Huangshi shifted from “under-resourced” to “under-utilized,” while Xiantao, Tianmen, and Qianjiang showed the opposite change.

Discussion

Potential access to healthcare services

As discussed in previous sections, this study utilized the standard deviational ellipse, average nearest neighbor, kernel density, and buffer analysis to measure the fit between medical and residential spaces (i.e., potential healthcare accessibility) in Wuhan Metropolitan Area. First, there is an overall spatial alignment between healthcare facilities and residential areas throughout the entire region, which implies the capacity of WMA’s healthcare services to satisfy most residents’ needs. However, the spatial analysis reveals an uneven potential healthcare accessibility in different areas. The distribution of medical and residential spaces is typically most coordinated in urban central areas, especially that of Wuhan. In contrast, healthcare facilities are inadequate in the urban–rural continuum, but in surplus in peripheral areas of some cities, such as Wuhan, Xiantao, Xianning, and Huanggang.

Moreover, potential spatial accessibility to healthcare facilities is also notably uneven across the nine cities. Wuhan residents enjoy the highest accessibility to healthcare facilities, whether primary facilities such as clinics and community health centers, or higher-level facilities such as general hospitals. This undoubtedly is closely connected with Wuhan’s dominant position in the urban hierarchy. However, the imbalance between provision and utilization in other cities of WMA is worth noting. While higher-level healthcare facilities in these cities manage to serve over 90% of the residential areas within a 15-minute walking distance, only approximately 70% of the settlements in cities except for Qianjiang are within a five-minute walk of primary-level healthcare facilities. This underscores that primary healthcare resources in WMA are far from adequate.

Realized access to healthcare services

Coupling coordination analysis is employed to understand the actual fit between healthcare services offered and utilized (i.e., residents’ realized access to healthcare services). Our analysis shows that despite the differences among the nine cities, the provision and utilization of medical resources in WMA as a whole is out of coordination. Wuhan has the highest coupling coordination degree of healthcare service provision and utilization, which reached quality coordination (V1) in 2019 but fell to intermediate coordination (V2) in 2020, indicating a generally satisfactory accessibility to healthcare services. Huangshi and Ezhou reachedat least primary coordination (V3) as well. However, all other cities have exhibited incoordination in the provision and utilization of healthcare resources to varying degrees at different times.

Combining the two aspects of coupling coordination degree, i.e., the level and the type, allows us to make observations about the impact caused by the COVID-19 pandemic. Among the nine cities, only Wuhan has been attracting migrants from other cities, particularly within Hubei province, while all the other eight cities have suffered from population loss. The COVID-19 pandemic temporarily changed this situation. It is noticeable that the coupling coordination degree in Wuhan exhibited a “utilization lag” in 2019; the COVID-19 outbreak and subsequent lockdowns in 2020 further magnified the underutilization of healthcare services. In contrast, Xiaogan, Huanggang, and Xianning continuously experienced insufficient healthcare resources, which was exacerbated in 2020. On the one hand, this can be attributed to their slow urban and economic development. Although these three cities are prefectural-level cities with a large population and administrative areas, their economic development (GDP per capita, etc.) was lagged, even compared with Xiantao, Qianjiang, and Tianmen, which are county-level cities (Hubei Provincial Statistics Bureau, 2021). On the other hand, the COVID-19 pandemic and the Wuhan lockdown required residents to stay put in their hometown and seek local healthcare services. This may also explain the shifting from “utilization lag” to “provision lag” in Xiantao, Tianmen, and Qianjiang. These three cities previously had a high number of out-migrants; but Wuhan’s lockdown and the COVID-19 pandemic blocked out-migration, and accordingly put great pressure on the local healthcare system.

Conclusion and implications

Healthcare accessibility is an important livelihood issue directly related to people’s health. China’s healthcare service capacity has improved significantly; however, the lack of coordination between provision and utilization of healthcare services remains prominent (Wang et al., 2019; Yin et al., 2018). By synthesizing spatial analysis and statistical modelling, this paper has revealed the healthcare accessibility in Wuhan Metropolitan Area, in both potential and realized dimensions. The distribution of healthcare facilities in WMA presents a clustering pattern, with a single and strong core emerging in the city center of Wuhan. Assuming a walking distance of 5 minutes for primary healthcare and 15 minutes for higher-level healthcare, we found residents’ potential access to higher-level healthcare facilities satisfactory; however, access to primary-level healthcare facilities still needs to be improved in most cities. Moreover, potential access to healthcare facilities is uneven across WMA’s nine cities and different areas. We call for more attention to areas in the urban–rural continuum, where healthcare facilities are most inadequate. In terms of the actual fit between healthcare services offered and utilized, our findings showed that except for Wuhan, all the other cities suffer from different levels of incoordination and therefore, a lack of resilience to public health emergencies.

The major contribution of this study is a combined evaluation of residents’ potential and realized access to healthcare services. The findings offer valuable information for policymakers to enhance the overall capacity of the healthcare system. First, the unevenness of economic development and healthcare resource allocation among different cities and regions suggests that governments at all scales in the metropolitan area should collaborate to formulate a more coordinated and comprehensive plan for resource allocation, and to promote resource sharing. On the one hand, it is crucial to enhance Wuhan’s radiation ability, since the core city of the metropolitan area enjoys the most advantaged healthcare services. Moreover, this paper reveals a surplus provision of healthcare resources in Wuhan. Hence, encouraging residents in the surrounding areas to make use of Wuhan’s medical resources can clearly improve their utilization efficiency. On the other hand, there is an urgent need to improve the healthcare resource abundance in cities such as Xiaogan, Huanggan, and Xianning, where the supply of healthcare services is not enough to meet residents’ demand. Second, both the quantity and quality of healthcare services need to be improved to enhance urban resilience to public health emergencies. Hit by the COVID-19 pandemic in 2020, the coupling coordination degree of healthcare service provision and utilization in all cities dropped, to varying extents. Moreover the number of tertiary hospitals and health technicians, were found to be more important than usual during the pandemic. Therefore, the regional and city governments should take the public health emergency into consideration to build a more resilient healthcare system.

The analysis is not without its shortcomings. One problem is the heterogenous healthcare accessibility issues faced by Chinese urban and rural residents. In terms of the provision of healthcare resources, the Principle of Medical Institution Planning issued by the National Health Commission highlights two parallel three-tier healthcare service systems for urban and rural regions. On the demand side, the rural population considers primary healthcare facilities (i.e., township health centers) as their default choice while urban residents prefer higher-level facilities (Liu et al., 2018). Secondly, this study evaluates the availability and utilization of healthcare services solely at the city level. Consequently, it overlooks the nuanced distinctions in healthcare facility types and levels, as well as different groups’ specific healthcare needs. These questions could be better addressed in future research with micro-level data.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China [42171205]

Notes

Author biographies

Sainan Lin is an Associate Professor in School of Urban Design at Wuhan University. She focuses on migration, residential mobility, social integration, and socio-spatial transformation in Chinese cities. Her work has also been published in Urban Studies, Urban Geography, Cities, Geoforum, and Habitat International.

Huilin Deng is a postgraduate student at the School of Urban Design at Wuhan University. She focuses on the development and governance innovation of China’s metropolitan areas, social integration, population migration and mobility.

Yu Wang is a Ph.D. student in the Edward J. Bloustein School of Planning and Public Policy at Rutgers University. She received her B.Eng. in Urban planning from Wuhan University, China, and her M.S. in Regional Science from Cornell University. She is interested in immigration and urban governance.

Si Chen is an economist at Wuhan Personnel Testing Authority. She focuses on employment of university graduates, doctor and social worker recruitment, licensed pharmacist examination and their impact on population movements, social mobility and economic growth.

References

Aday

Andersen

(1974) A framework for the study of access to medical care. Health Services Research 9(3): 208–220.

Andersen

(1995) Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior 36(1): 1–10.

Andersen

Davidson

(2007) Improving access to care in America: Individual and contextual indicators. In: Andersen

Rice

Kominski

(eds) Changing the US Health Care System: Key Issues in Health Services Policy and Management. San Francisco, CA: Jossey-Bass, 3–31.

Andersen

Newman

(1973) Societal and individual determinants of medical care utilization in the United States. The Milbank Memorial Fund Quarterly. Health and Society 51(1): 95–124.

Andritsos

Tang

(2014) Introducing competition in healthcare services: The role of private care and increased patient mobility. European Journal of Operational Research 234(3): 898–909.

Ashik

Mim

Neema

(2020) Towards vertical spatial equity of urban facilities: An integration of spatial and aspatial accessibility. Journal of Urban Management 9(1): 77–92.

Cartier

(2004) City-space: Scale relations and China’s spatial administrative hierarchy. In: Ma

LJC

(eds) Restructuring the Chinese City. London: Routledge, 19–33.

Chen

, et al. (2022) Different functional areas and human activities significantly affect the occurrence and characteristics of microplastics in soils of the Xi’an metropolitan area. Science of the Total Environment 852: 158581.

Chen

Z-L

Zhang

, et al. (2020a) Distribution of the COVID-19 epidemic and correlation with population emigration from Wuhan, China. Chinese Medical Journal 133(9): 1044.

10.

Chen

Sleipness

, et al. (2020b) A systematic review of alternative protocols for evaluating non-spatial dimensions of urban parks. Urban Forestry & Urban Greening 53: 126718.

11.

Coleman

(1975) Problems of conceptualization and measurement in studying policy impacts. In: Dolbeare

(ed.) Public Policy Evaluation, vol. 2. Beverly Hills, CA: Sage, 19–40.

12.

Dalvi

Martin

(1976) The measurement of accessibility: Some preliminary results. Transportation 5(1): 17–42.

13.

Dejen

Soni

Semaw

(2019) Spatial accessibility analysis of healthcare service centers in Gamo Gofa Zone, Ethiopia through Geospatial technique. Remote Sensing Applications: Society and Environment 13: 466–473.

14.

Ding

Yang

Wang

, et al. (2023) Influential factors of intercity patient mobility and its network structure in China. Cities 132: 103975.

15.

Donabedian

(1966) Evaluating the quality of medical care. The Milbank Memorial Fund Quarterly 44(3): 166–206.

16.

Donabedian

(1973) Aspects of Medical Care Administration: Specifying Requirements for Health Care. Cambridge, MA: Harvard University Press.

17.

Fukuda

(2020) Land prices and agglomeration: Theory and evidence from the Tokyo metropolitan area. Journal of the Japanese and International Economies 58: 101094.

18.

Gao

Wang

, et al. (2020) A correction to the entropy weight coefficient method by Shen et al. for accessing urban sustainability [Cities 42 (2015) 186–194]. Cities 103: 102742.

19.

Geurs

Van Wee

(2004) Accessibility evaluation of land-use and transport strategies: Review and research directions. Journal of Transport Geography 12(2): 127–140.

20.

Ghorbanzadeh

Kim

Erman Ozguven

, et al. (2021) Spatial accessibility assessment of COVID-19 patients to healthcare facilities: A case study of Florida. Travel Behaviour and Society 24: 95–101.

21.

Giles-Corti

Vernez-Moudon

Reis

, et al. (2016) City planning and population health: A global challenge. The Lancet 388(10062): 2912–2924.

22.

Guagliardo

(2004) Spatial accessibility of primary care: Concepts, methods and challenges. International Journal of Health Geographics 3(1): 3.

23.

Guida

Carpentieri

(2021) Quality of life in the urban environment and primary health services for the elderly during the Covid-19 pandemic: An application to the city of Milan (Italy). Cities 110: 103038.

24.

Haldane

De Foo

Abdalla

, et al. (2021) Health systems resilience in managing the COVID-19 pandemic: Lessons from 28 countries. Nature Medicine 27(6): 964–980.

25.

Hawthorne

Kwan

M-P

(2013) Exploring the unequal landscapes of healthcare accessibility in lower-income urban neighborhoods through qualitative inquiry. Geoforum 50: 97–106.

26.

Haynes

Lovett

Sünnenberg

(2003) Potential accessibility, travel time, and consumer choice: Geographical variations in general medical practice registrations in Eastern England. Environment and Planning A: Economy and Space 35(10): 1733–1750.

27.

Zhang

(2022) The distribution effect of urbanization: Theoretical deduction and evidence from China. Habitat International 123: 102544.

28.

Huang

(2008) Combining entropy weight and TOPSIS method for information system selection. In: IEEE conference on cybernetics and intelligent systems, Chengdu, China, September, 1281–1284. New York, NY: IEEE.

29.

Huang

Meyer

Jin

(2019) Spatial access to health care and elderly ambulatory care sensitive hospitalizations. Public Health 169: 76–83.

30.

Hubei Provincial Statistics Bureau (2021) Hubei Statistical Yearbook 2021. Wuhan, China: China Statistics Press.

31.

Ingram

(1971) The concept of accessibility: A search for an operational form. Regional Studies 5(2): 101–107.

32.

Joseph

Bantock

(1982) Measuring potential physical accessibility to general practitioners in rural areas: A method and case study. Social Science & Medicine 16(1): 85–90.

33.

Joseph

Phillips

(1984) Accessibility and Utilization: Geographical Perspectives on Health Care Delivery. London: Harper & Row.

34.

Kanuganti

Sarkar

Singh

(2016) Quantifying accessibility to health care using Two-step Floating Catchment Area Method (2SFCA): A case study in Rajasthan. Transportation Research Procedia 17: 391–399.

35.

Khan

(1992) An integrated approach to measuring potential spatial access to health care services. Socio-Economic Planning Sciences 26(4): 275–287.

36.

Wang

(2022) A hierarchical two-step floating catchment area analysis for high-tier hospital accessibility in an urban agglomeration region. Journal of Transport Geography 102: 103369.

37.

Yuan

Qin

, et al. (2022) Coupling coordination degree for urban green growth between public demand and government supply in urban agglomeration: A case study from China. Journal of Environmental Management 304: 114209.

38.

Rivers

Tan

, et al. (2020) Estimated demand for US hospital inpatient and intensive care unit beds for patients with COVID-19 based on comparisons with Wuhan and Guangzhou, China. JAMA Network Open 3(5): e208297.

39.

Zhou

, et al. (2012) Investigation of a coupling model of coordination between urbanization and the environment. Journal of Environmental Management 98: 127–133.

40.

Lin

S-J

Crawford

Warren Salmon

(2005) Potential access and revealed access to pain management medications. Social Science & Medicine 60(8): 1881–1891.

41.

Liu

Zhong

Yuan

, et al. (2018) Why patients prefer high-level healthcare facilities: A qualitative study using focus groups in rural and urban China. BMJ Global Health 3(5): e000854.

42.

Luo

Wang

(2003) Measures of spatial accessibility to health care in a GIS environment: Synthesis and a case study in the Chicago region. Environment and Planning B: Planning and Design 30(6): 865–884.

43.

(2022) Evolutionary networks of interurban technological collaboration across Chinese city-regions. Regional Studies, Regional Science 9(1): 457–460.

44.

McGrail

(2012) Spatial accessibility of primary health care utilising the two step floating catchment area method: An assessment of recent improvements. International Journal of Health Geographics 11(1): 50.

45.

McLafferty

(1982) Urban structure and geographical access to public services. Annals of the Association of American Geographers 72(3): 347–354.

46.

Martin

Wrigley

Barnett

, et al. (2002) Increasing the sophistication of access measurement in a rural healthcare study. Health & Place 8(1): 3–13.

47.

Penchansky

Thomas

(1981) The concept of access: Definition and relationship to consumer satisfaction. Medical Care 19(2): 127–140.

48.

Rahman

Ashik

Mouli

(2022) Investigating spatial accessibility to urban facility outcome of transit-oriented development in Dhaka. Transportation Research Interdisciplinary Perspectives 14: 100607.

49.

Ren

Tarimo

Sun

, et al. (2021) The degree of equity and coupling coordination of staff in primary medical and health care institutions in China 2013–2019. International Journal for Equity in Health 20(1): 236.

50.

Shadmi

Wong

Kinder

, et al. (2014) Primary care priorities in addressing health equity: Summary of the WONCA 2013 health equity workshop. International Journal for Equity in Health 13(1): 104.

51.

Shannon

Spurlock

Gladin

, et al. (1975) A method for evaluating the geographic accessibility of health services. The Professional Geographer 27(1): 30–36.

52.

Shen

Sun

(2022) Estimating the spatial correlation and convergence of China’s healthcare resources allocation: Evidence from the Yangtze River Delta Region. Archives of Public Health 80(1): 207.

53.

Sherman

Spencer

Preisser

, et al. (2005) A suite of methods for representing activity space in a healthcare accessibility study. International Journal of Health Geographics 4(1): 24.

54.

Shi

Yang

Zhang

, et al. (2020) Coupling coordination degree measurement and spatiotemporal heterogeneity between economic development and ecological environment——Empirical evidence from tropical and subtropical regions of China. Journal of Cleaner Production 244: 118739.

55.

Spencer

Angeles

(2007) Kernel density estimation as a technique for assessing availability of health services in Nicaragua. Health Services and Outcomes Research Methodology 7(3): 145–157.

56.

Sun

Miao

Yang

(2017) Ecological-economic efficiency evaluation of green technology innovation in strategic emerging industries based on entropy weighted TOPSIS method. Ecological Indicators 73: 554–558.

57.

Taleai

Sliuzas

Flacke

(2014) An integrated framework to evaluate the equity of urban public facilities using spatial multi-criteria analysis. Cities 40: 56–69.

58.

Talen

Anselin

(1998) Assessing spatial equity: An evaluation of measures of accessibility to public playgrounds. Environment and Planning A: Economy and Space 30(4): 595–613.

59.

Tao

Yao

Kong

, et al. (2018) Spatial accessibility to healthcare services in Shenzhen, China: Improving the multi-modal two-step floating catchment area method by estimating travel time via online map APIs. BMC Health Services Research 18(1): 345.

60.

Wan

(2022) High-speed railway and the intercity transmission of epidemics: Evidence from COVID-19 in China. Economic Modelling 114: 105934.

61.

Wang

(2012) Measurement, optimization, and impact of health care accessibility: A methodological review. Annals of the Association of American Geographers 102(5): 1104–1112.

62.

Wang

(2018) Inverted two-step floating catchment area method for measuring facility crowdedness. The Professional Geographer 70(2): 251–260.

63.

Wang

Luo

(2005) Assessing spatial and nonspatial factors for healthcare access: Towards an integrated approach to defining health professional shortage areas. Health & Place 11(2): 131–146.

64.

Wang

, et al. (2019) The development and reform of public health in China from 1949 to 2019. Globalization and Health 15(1): 45.

65.

Weiss

Nelson

Gibson

, et al. (2018) A global map of travel time to cities to assess inequalities in accessibility in 2015. Nature 553(7688): 333–336.

66.

Wen

Chatalova

Gao

, et al. (2021) Reduction of carbon emissions through resource-saving and environment-friendly regional economic integration: Evidence from Wuhan metropolitan area, China. Technological Forecasting and Social Change 166: 120590.

67.

World Health Organization (2020) WHO COVID-19 explorer. Available at: https://worldhealthorg.shinyapps.io/covid/ (accessed 18 January 2023).

68.

Wuhan Municipal Bureau of Statistics (2021a) Main data of the seventh national population census in Wuhan. Available at: http://tjj.wuhan.gov.cn/ztzl_49/pczl/202109/t20210916_1779157.shtml (accessed 18 January 2023).

69.

Wuhan Municipal Bureau of Statistics (2021b) Interpretation of main data of the seventh national population census in Wuhan. Available at: http://news.cjn.cn/sywh/202105/t3829405.htm?spm=zm1066-001.0.0.1.rciJC6 (accessed 18 January 2023).

70.

Xinhua News Agency (2014) National new urbanization plan (2014–2020). Available at: http://www.gov.cn/zhengce/2014-03/16/content_2640075.htm (accessed 18 January 2023).

71.

Xinhua News Agency (2020) Xinhua Headlines: Wuhan reopens after 76-day life-and-death battle against novel coronavirus. Available at: http://www.xinhuanet.com/english/2020-04/08/c_138958718.htm (accessed 18 January 2023).

72.

Yang

D-H

Goerge

Mullner

(2006) Comparing GIS-based methods of measuring spatial accessibility to health services. Journal of Medical Systems 30(1): 23–32.

73.

Yin

Liu

, et al. (2018) Inequality of public health and its role in spatial accessibility to medical facilities in China. Applied Geography 92: 50–62.

74.

Zeng

Liu

Stein

, et al. (2015) Characterization and spatial modeling of urban sprawl in the Wuhan Metropolitan Area, China. International Journal of Applied Earth Observation and Geoinformation 34: 10–24.

75.

Zhang

Wang

Jiang

, et al. (2018) Equity and efficiency of primary health care resource allocation in mainland China. International Journal for Equity in Health 17(1): 140.

76.

Zhang

, et al. (2022) Spatio-temporal characteristics of the supply and demand coupling coordination of elderly care service resources in China. International Journal of Environmental Research and Public Health 19(16): 10397.

77.

Zhu

Jiao

Zhang

, et al. (2021) Sustainable development of urban metro system: Perspective of coordination between supply and demand. International Journal of Environmental Research and Public Health 18(19): 10349.

78.

Zhu

Tian

Yan

(2020) Effectiveness of entropy weight method in decision-making. Mathematical Problems in Engineering 2020: 1–5.

79.

Zhuang

Cao

Zhao

, et al. (2021) The shortage of hospital beds for COVID-19 and non-COVID-19 patients during the lockdown of Wuhan, China. Annals of Translational Medicine 9(3): 200.