EN中文

Journal of Resources and Ecology >

2024 , Vol. 15 >Issue 1: 130 - 139

DOI: https://doi.org/10.5814/j.issn.1674-764x.2024.01.011

Resource Economy

Spatial Distribution and Factors Influencing Elderly Care Service Facilities based on Accessibility Evaluation—Taking Wuxi City as an Example

  • ZHAO Li , 1 ,
  • HE Diaoxia 1 ,
  • ZHAO Jian 2
Expand
  • 1. Wuxi Vocational Institute of Commerce, Wuxi, Jiangsu 214153, China
  • 2. Educational Technology School, Northwest Normal University, Lanzhou 730070, China

Received date: 2023-06-20

  Accepted date: 2023-08-30

  Online published: 2023-12-04

Supported by

The Research Project on Philosophy and Social Sciences in Jiangsu Universities and Colleges(2020SJA0933)

The 2022 Tourism Vocational Education Research Project of Jiangsu Tourism Vocational Education Group(HT2023025)

The 2022 “Double High Construction” Special Project of Wuxi Vocational Institute of Commerce(SGZXGJ20220109)

Fold

Abstract

With the progressive aging of China’s population, the contradiction between the supply and demand of elderly care service facilities is increasingly prominent. This study takes Wuxi City as the research area, and evaluates the accessibility characteristics of the elderly service facilities in Wuxi City by using geographic statistics, buffer analysis, Gaussian two-step mobile search and other methods. It then analyzes the spatial distribution and supply and demand of the elderly service facilities in Wuxi City, and introduces the XGBoost algorithm to calculate the factors influencing the distributions of different types of elderly services. The results show that the distribution of home-based and institutional endowment resources in Wuxi is spatially heterogeneous, but there are problems of mismatch between supply and demand and uncoordinated regional development. The accessibility of endowment services varies greatly in different regions, and the number of hierarchical endowment institutions is relatively small. The distribution of community home care service stations is relatively more equitable, but still does not match the size of the elderly population. There are obvious differences in the factors affecting the distributions of nursing institutions and community home nursing service stations. The distribution of nursing institutions is affected by transportation infrastructure, the size of the elderly population, the number of medical resources, the environment, economic level and other factors; while community home care is mainly affected by the size of the elderly population. In order to better meet the elderly service needs of residents, Wuxi needs to increase its investment in elderly service facilities and take various measures to promote the development and improvement of its elderly service facilities.

Key words： accessibility evaluation; Gauss two-step mobile search method; XGBoost; elderly care service; Wuxi City

Cite this article

ZHAO Li , HE Diaoxia , ZHAO Jian . Spatial Distribution and Factors Influencing Elderly Care Service Facilities based on Accessibility Evaluation—Taking Wuxi City as an Example[J]. Journal of Resources and Ecology, 2024 , 15(1) : 130 -139 . DOI: 10.5814/j.issn.1674-764x.2024.01.011

1 Introduction

Statistics from the seventh national census in 2020 show that 18.7% of China’s population is aged 60 or above, and the degree of aging is still growing rapidly. In Jiangsu Province, 21.84% of its permanent residents are aged over 60 years old, so it is one of the most prominent areas in China with an aging problem. With the acceleration of the aging process, the contradiction between the supply and demand of elderly care service facilities has become increasingly prominent. In addition to an insufficient supply of elderly care facilities, there are also problems with the dislocation of supply and demand in space and uncoordinated regional development (Han and Luo, 2020). The spatial configuration of pension service facilities is related to whether the elderly pension services can be effectively guaranteed.
Spatial accessibility evaluation is an effective method for identifying the spatial configuration of public service facilities. It refers to the convenience of residents in overcoming resistance factors from one place to another place, so it reflects the spatial distribution of potential destinations, the difficulty in obtaining opportunities, and the scale or attraction of activities in a specific place. It is an important index for measuring the rationality and fairness of the spatial distribution of urban service facilities (Handy and Niemeier, 1997; Wang, 2012), and a research hotspot in the field of spatial geography. There are three ways to measure the spatial accessibility of services for elderly care facilities. One is based on the spatial partition, which is measured by the number of services and the per capita number in the spatial area or administrative unit (Ding et al., 2022; Li et al., 2023). The second is distance-based measurement, which calculates the population within a certain spatial distance or commuting time from service facilities, mainly using the shortest distance method (Zhang and Zhang, 2016). The third method is gravity-based measurement, which comprehensively considers the quantity of service supply, scale of population demand, and spatial distance or commuting time, including the potential model (Ding et al., 2016), two-step mobile search method (Wang et al., 2023), and others. Comparatively, the first two methods are more scientific. The difference between the two-step mobile search method and the gravity model method lies in their treatment of distance factors. The gravity model method adopts the continuous distance attenuation function, which considers the attenuation of service capability with distance, but does not limit the effective radius of service. On the other hand, the two-step mobile search method adopts the dichotomy method to deal with distance attenuation, so it has the same reachability within the search radius threshold, but it is completely inaccessible outside the search radius (Tao and Cheng, 2016).
Many domestic studies have analyzed the accessibility of elderly care services and reported many achievements (Zhao et al., 2014; Xu and Zhao, 2017; Xu and Tu, 2020; Yin and Liu, 2021; Zhang et al., 2023). However, from the perspective of research areas, the current domestic analysis on the accessibility of urban elderly care resources has mainly focused on metropolises such as Beijing (Zhang et al., 2023), Shanghai (Ding et al., 2016; Tao and Cheng, 2016; Zhang and Zhang, 2016; Yin and Liu, 2021; Wang et al., 2023; Zhang et al., 2023), Wuhan (Ding et al., 2016; Ding et al., 2022; Li et al., 2023) and Nanjing (Xu and Zhao, 2017), while other studies have focused on sub-provincial or provincial capitals such as Dalian (Zhao et al., 2014) and Fuzhou (Xu and Tu, 2020). However, scholars still pay only limited attention to the study of prefecture-level cities. As a prefecture-level city with a relatively developed economy in the Yangtze River Delta, Wuxi is one of the earliest cities to become part of the aging society in China. It also has a relatively high degree of population aging in China, and faces the severe problem of population aging. Therefore, the empirical analysis of the accessibility of supporting resources for the elderly in Wuxi is of great significance for improving and optimizing the distribution of Wuxi’s supporting resources for the elderly. From the perspective of research content, existing studies mainly focused on evaluating the accessibility and analyzing the supply-demand characteristics of elderly care resources, but paid little attention to the factors influencing the distribution of elderly care service facilities.
In this study, road network data, population raster data, elderly care institution and community home care service station data were used to evaluate the accessibility characteristics of elderly care service facilities in Wuxi City through zoning statistics, buffer analysis, the Gauss two-step mobile search and other methods, in order to determine the relationship between the supply and demand of elderly care resources. The Gini coefficient was used to calculate the fairness of the spatial distribution of elderly care resources. Moreover, geographical data such as luminous remote sensing and poi were used to introduce the XGBoost algorithm in order to evaluate the importance of the various influencing factors on the distribution of elderly care institutions and community home care service stations. The results of this study provide references for decision-making and promote the development of high-quality elderly care services.

2 Research area, data and methods

2.1 Research area

The study takes Wuxi City of Jiangsu Province as the research area. Wuxi is located in the Yangtze Delta Economic Zone, adjacent to the Taihu Lake, and includes seven districts and counties. According to the data of the 7th national population census, the aging degree of the Yangtze River Delta region was higher than the national average level, and it had entered the stage of moderate population aging. According to the 7th national census bulletin of Jiangsu Province, the permanent elderly population aged 60 and above in Jiangsu Province was 18.5053 million, accounting for 21.84% of the total population, ranking sixth in the country and second only to Shanghai in the Yangtze River Delta region. As the central city of Southern Jiangsu, the population of Wuxi aged 60 and above was 1.47 million, accounting for 19.75% of the total population; while the population aged 65 and above was 1.09 million, accounting for 14.65%. Compared with the sixth national population census in 2010, the proportion of the population aged 60 and above in Wuxi increased by 5.05 percentage points, and the proportion aged 65 and above increased by 5.17 percentage points in 2020. As one of the most prominent cities with aging population problems in Jiangsu Province, Wuxi has entered the stage of deep aging. Therefore, compared with other regions in China, there is a more urgent need for developing the elderly care industry in Wuxi City.

2.2 Data sources

The population data came from worldpop, and the 100-m resolution raster data of the age-specific population of Wuxi in 2020 was obtained and trimmed. The data for the elderly over 60 years old were extracted and corrected by referring to the data of the Jiangsu Provincial Bureau of Statistics. Road network and poi data were from Amap. The data of elderly care institutions came from the website of Wuxi Civil Affairs Bureau and the elderly Care Information network (http://www.yanglaocn.com/yanglaoyuan/yly/), which include data on 163 elderly care institutions and 1144 community home care service stations.

2.3 Research methods

2.3.1 Gauss two-step mobile search method

The two-step mobile search method was first proposed by Radke and Mu (2000), with the basic idea as follows.
For each elderly care service facility point j, set a service threshold d0, search for the elderly population demand point k within the search radius d0, and calculate the supply and demand ratio Rj. The supply and demand ratio Rj represents the matching degree between supply point j and demand point k.
${{R}_{j}}=\frac{{{s}_{j}}}{{{\Sigma }_{k\in \left\{ {{d}_{kj\le }}{{d}_{0}} \right\}}}{{D}_{k}}}$
In formula (1), j denotes the supply point, k represents the demand point, d0 is the service threshold, dk j is the distance from demand point k to supply point j, sj is the number of beds at point j, and Dk is the total demand value at point k.
For each demand point i, search for the supply points within its search radius d0, and sum their supply and demand ratios Rj to obtain the accessibility $A_{i}^{F}$ of all the points.
$A_{i}^{F}={{\Sigma }_{i\in \left\{ {{d}_{ij\le }}{{d}_{0}} \right\}}}{{R}_{j}}$
In formula (2), the accessibility $A_{i}^{F}$ indicates the closeness between the demand point i and its surrounding supply point j, i denotes the demand point, d0 represents the service threshold, dij is the distance from demand point i to supply point j, and Rj is the supply-demand ratio of elderly care service facility j.
Dai proposed the Gauss two-step mobile search method (Gaussian 2SFCA) (Dai, 2010). It uses a Gaussian function as the distance attenuation function $g({{d}_{kj}},{{d}_{0}})$ within the search radius of 2SFCA, which can be expressed as:
$g\left( {{d}_{kj}},{{d}_{0}} \right)=\left\{ \begin{matrix} \frac{{{\text{e}}^{-\frac{1}{2}\times {{\left( \frac{{{d}_{kj}}}{{{d}_{0}}} \right)}^{2}}}}-{{\text{e}}^{-\frac{1}{2}}}}{1-{{\text{e}}^{-\frac{1}{2}}}},{{d}_{kj}}\le {{d}_{0}} \\ 0,{{d}_{kj}}>{{d}_{0}} \\ \end{matrix} \right.$
In formula (3), d0 denotes the service threshold, and dkj represents the distance from the demand point k to the supply point j.
Using a Gaussian function as the search radius of 2SFCA, the accessibility of elderly care institutions can be expressed as:
$A_{i}^{F}=g({{d}_{kj}},{{d}_{0}})\times {{\Sigma }_{i\in \left\{ {{d}_{ij\le }}{{d}_{0}} \right\}}}{{R}_{j}}$
In formula (4), $g({{d}_{kj}},{{d}_{0}})$ is the Gaussian equation function that considers the distance attenuation effect.
The Gaussian two-step mobile search method was used to calculate the accessibility of beds in elderly care institutions. The driving speeds of expressways, trunk roads and branch roads were set as 35, 25 and 21 km h-1, respectively, and the remaining lanes were set as 16 km h-1. According to The Construction Guide of Standard System of Pension Service (2017), the scale of elderly care institutions is divided into four categories: micro, small, medium and large. The number of beds in micro elderly care institutions is less than 60, and the maximum service radius is 0.5 h of vehicle traffic distance. The number of beds in small elderly care institutions ranges from 61 to 120, and the maximum service path is 0.9 h. The number of beds of medium elderly care institutions is 121-220, and the maximum service radius is 1.2 h. The number of beds in large elderly care institutions reaches more than 220, and the maximum service radius is 1.5 h.

2.3.2 XGBoost algorithm

XGBoost (Extreme Gradient Boosting), a machine learning algorithm based on the gradient Boosting framework (Chen and Guestrin, 2016), is widely used in data science and predictive modeling. It is an ensemble learning method for improving the predictive performance by serial training and the integration of multiple weak learners. It works through three steps. 1) It builds an initial weak learner, usually a decision tree model, to make predictions about the data. 2) It calculates the residual between the predicted result and the actual result, and takes the residual as the training target of the next weak learner. 3) It then uses a gradient lifting algorithm to train more weak learners gradually and iteratively, and adds them to the integrated model. XGBoost introduces regularization terms in the objective function to control the complexity of the model and avoid overfitting, with a regularization coefficient set to 0.2. The importance is calculated based on the average information gain of features in the model, and the formula for calculating information gain is as follows:
$Gain=\frac{1}{2}[\frac{G_{L}^{2}}{{{H}_{L}}+\lambda }+\frac{G_{R}^{2}}{{{H}_{R}}+\lambda }+\frac{{{({{G}_{L}}+{{G}_{R}})}^{2}}}{{{H}_{L}}+{{H}_{R}}+\lambda }]-\gamma$
In formula (5), GL and GR represent the gradient statistics of the left and right child nodes, respectively; that is, the first derivative of the loss function; HL and HR represent the Hessian statistics of the left and right child nodes, respectively; that is, the second derivative of the loss function; λ is the regularization parameter; γ is the regularization parameter of the leaf node.

3 Research results

The distribution of the elderly population over 60 years old in Wuxi presents obvious heterogeneity. The urban area is concentrated in Liangxi District, the northern part of Jiangyin City is banded along the river, and Yixing City is represented by several agglomeration centers. Jiangyin and Yixing have a large elderly population, but the density of the county is low, and the distribution of the elderly population is more dispersed. The elderly population in Liangxi District has the highest density and the most concentrated distribution. Chong’an Temple Subdistrict, South Zen Temple Street, Tongjiang Street, and Shang Madun Street in Liangxi District are the most densely populated areas. There are nine streets (towns) with elderly populations exceeding 30000, which are Dingshu Town, Yicheng Street and Zhangzhu Town in Yixing City, Chengjiang Street and Lingang Street in Jiangyin City, Zhouzhuang Town, Jiangyin High-tech Industrial Development Zone, Huashi Town and Jiangxi Street in Xinwu District. Among them, the number of elderly people is more than 60000 in Chengjiang Street in Jiangyin City and Dingshu Town in Yixing City.
The bed density of elderly care institutions declines from Binhu District and Liangxi District to the surrounding area, and its distribution is strongly correlated with the distribution of the elderly population (Fig. 1). Liangxi District and Binhu District have the most densely distributed elderly care institutions, and most large elderly care institutions are concentrated there. Elderly care institutions in Huishan District, Xishan District, Yixing City and Jiangyin City are relatively sparse, and Huishan District, Yixing City and Jiangyin City do not have any large elderly care institutions. The distribution of elderly care institutions does not match the distribution of the elderly population. Areas such as high-tech industrial development zone, Yanqiao Street, Huishan Economic Development Zone, Helie Street, Meicun Street, Shanbei Street, North Street, Nanzha Street, Hongshan Street, and Yuqi Streets have large elderly populations, but lack any elderly care institutions. Elderly care institutions in Zhouzhuang Town, Jiangyin High-tech Industrial Development Zone, Xishan Development Zone, Guanlin Town, Huangxiang Sub-district, Xuxiake Town and other areas are relatively scarce, and the number of beds in elderly care institutions is less than 0.2 per 100 people. In contrast, Rongxiang Sub-district, Taihua Town, Guangrui Road sub-district and Liyuan sub-district are relatively rich in elderly care resources, with the number of beds for every 100 people exceeding 8. Generally, the distribution of elderly care institutions is very different in different streets, which does not match the size of elderly population, and the Gini coefficient is more than 0.9. However, the distribution of elderly care institutions is often not determined by the singular factor of the distribution of the elderly population, so a variety of factors need to be considered comprehensively. Therefore, it is more reasonable to use the Gaussian two-step mobile search method for the accessibility analysis.
Fig. 1 Distribution of beds in elderly care institutions in Wuxi City
The bed accessibility per capita was calculated by the two-step mobile search method (Fig. 2). Overall, 66 towns or subdistricts have more than one bed for every 100 people. The average bed accessibility values per 100 people in Meishan Town, Hufu Town, Shuikou Town, Dingshu Town and Yicheng subdistrict are lower than 0.4, so they are each at a low level. Taihua Town has the highest number of beds per 100 people at 3.48. From the perspective of different districts and counties, the accessibility of elderly care insti tution beds in Yixing City is extremely uneven. Taihua Town in the southwest end of Yixing City has the highest accessibility, and Wanshi Town in the northeast end of Yixing City has relatively good accessibility, but the accessibility values in other areas are relatively low. Most areas in Hufu Town and Dingshu Town have very low accessibility. Liangxi District, Binhu District and Xishan District have relatively high numbers of accessible beds in their elderly care institutions. The numbers of accessible beds in elderly care institutions are high in Mashan Street, Xuelang Street, Huazhuang Street and Wuxi Taihu International Science and Technology Park in the south of Binhu District, Yangjian Town, Ehu Town and Xishan Development Zone in the east of Xishan District. In Xinwu District, the accessibility is high in the west and low in the east. The accessibility of Xin’an Street and Wangzhuang Street is high in the west, while the accessibility of Shuofang Street and Hongshan street is relatively low in the east. The overall accessibility of elderly care institutions in Jiangyin City is low, and there is obvious spatial heterogeneity of accessibility, which is high in the south and low in the north, as well as high in the west and low in the east. The bed accessibility of elderly care institutions is low in Chengjiang Street, Nanzha Street, High-tech Industrial Development Zone, Zhouzhuang Town, Huashi Town and other areas.
Fig. 2 Bed accessibility of elderly care institutions in Wuxi City
The coverage range of the population under different radii of Grade 5 high-quality elderly care institutions is shown in Table 1. There are three Grade 5 elderly care institutions in Wuxi, namely Jinxi Yannian Leyi Nursing Home, Baihe Yiyang Nursing Home and Taihu Old-age Service Center. All of them are located in Binhu District, which are respectively located in Hudai Town, Rongxiang Street and Taihu Street. The distribution of large elderly care institutions may be affected by the distribution of the elderly population, land price and environment. For example, Hu Dai Town, Rongxiang Sub-district and Taihu Sub-district are not the most densely populated areas with respect to their elderly populations, but they are closer to the densely populated areas of the elderly population, and the environment is better. The Taihu Center serves the largest number of elderly people within the ranges of 500 m and 3 km. Within 1 km, Baihe Yiyang Nursing Home has the largest elderly population. Jinxi Yannian Leyi Nursing Home has the smallest elderly populations covered under the three distances. In addition to the Grade 5 elderly care institutions, there are also 18 Grade 4 elderly care institutions in Wuxi, but that number is still too low. High-quality elderly care resources are scarce, and the problem of spatial imbalance is prominent.
Table 1 The sizes of the elderly populations covered by Grade 5 elderly care institutions in Wuxi City
Elderly care
service facility		 500 m covers
the elderly
population		 1 km covers
the elderly
population		 3 km covers
the elderly population
Jinxi Yannian Leyi
Nursing Home		 466.63 13773.15 2124.68
Baihe Yiyang
Nursing Home		 654.95 20767.56 2538.54
Taihu Lake Senior
Care Service Center		 1023.87 18692.87 3623.82
The distribution of community home care service stations is shown in Fig. 3. The service radius of community home care service stations is smaller, so the per capita number of community home care service stations was calculated based on the street boundaries. There are obvious differences in the per capita resources of community home care service stations in different streets. A total of 16 streets (towns) had more than one community home care service station per 1000 people, and they were Dongting Street, Xuelang Street, Fangqiao Town, Yangxiang Town, Ehu Town, Xizhu Town, Zhoutie Town, Xinzhuang Street, Heqiao Town, Hufu Town, Taihua Town, Xushe Town, Xin’an Street, Xinjian Town, Gaocheng Town and Shuofang Street. Among them, the number of community home care service stations and the number of community home care service stations per thousand people in Dongting Street are the highest, at 362 and 21.98 per thousand, respectively. In Qingyang Town, Zhutang Town, Lihu Street, the number of community home care service stations per 1000 people is less than 0.1. Community home care service stations are usually non-profits, and the Gini coefficient of community home care service stations in different streets (towns) is 0.47, which is relatively fair compared with the distribution of elderly care institutions, but it still does not match the size of the elderly population.
Fig. 3 Distribution of home elderly care institutions in Wuxi City
The supply and demand values of elderly care resources in different districts and counties are shown in Table 2. The large elderly care institutions in Wuxi are mainly concentrated in the vicinity of Liangxi, which is close to areas with dense elderly populations. Binhu District has the largest number of beds in elderly care institutions, accounting for more than half of the total number of beds in the city, the number of beds per capita is the largest, and all Grade 5 high-quality elderly care institutions are located in Binhu District. Yixing has the largest number of community home care service stations, while Huishan, Liangxi and Yixing have the largest numbers of community home care service stations per capita, making up for the shortage of beds per capita. However, Xinwu District has the lowest number of beds in elderly care institutions and community home nursing service stations, and the supply of per capita elderly care facilities is relatively short. Jiangyin City has a large number of beds in elderly care institutions and community home nursing service stations, but due to the large size of the elderly population, the per capita elderly care resources are relatively insufficient.
Table 2 Supply-demand values of elderly care resources in different districts and counties
District/City Beds in elderly
care institutions		 Community home care service stations Elderly population (million) Beds per
100 people		 Home care service stations
in 1000-people community
Xishan 2805 105 0.13 2.21 0.83
Huishan 876 127 0.14 0.67 0.96
Binhu 12700 112 0.16 8.10 0.71
Liangxi 1442 160 0.18 0.83 0.92
Xinwu 950 73 0.16 0.60 0.46
Jiangyin 2002 266 0.36 0.57 0.75
Yixing 1948 301 0.34 0.59 0.91
Wuxi City (Total) 22723 1144 1.47 1.59 0.80
As for the factors affecting the distribution of elderly care resources, Jiang et al. reported that the number of elderly people, the level of economic development, public financial expenditure, the number of pension insurance participants and the area of green parks are the main factors affecting the number of urban elderly care institutions (Jiang et al., 2021). However, Han and Luo suggested that the spatial distribution of the elderly population, the urban traffic microcirculation system, facility environment and service quality are the main factors affecting the spatial differentiation of the supply and demand matching degree of elderly care facilities (Han and Luo, 2020). In another study, Gao found that the preference of elderly people for elderly care facilities with different features, locations, service standards and charges is significantly differentiated (Gao, 2013).
According to the published literature and the availability of data, this study selected six factors that may affect the distribution of elderly care resources, including the size of the elderly population, the level of economic development, the number of medical resources, the environment, the transportation infrastructure and the competition of different types of elderly care institutions; and then used the XGBoost algorithm to calculate the relative importance of these different factors. The sizes of the elderly populations in different regions in 2020 were obtained by the partition statistics of the 100-meter resolution raster data of Worldpop. Amap poi data were used to calculate the number of local medical resources. GEE was used to synthesize and correct the VIIRS data, and the average night light grid in 2020 was obtained to represent the level of local economic development. The shortest distance from the local area to the water body was calculated to represent the environmental factors. The traffic infrastructure was represented by calculating the road network length per unit area. The competition of different types of local elderly care services was represented by the number of home care services and institutional care services.
The feature importance values of the XGBoost measurements are shown in Fig. 4. There are obvious differences in the importance among the different factors for the distribution of the two kinds of elderly care service resources. The distribution of elderly care institutions is affected by transportation infrastructure, the size of the elderly population, the number of medical resources, the environment, economic level and other factors, among which the transportation infrastructure factor has the greatest impact, with an importance value of 0.23. While community home care is mainly affected by the size of the elderly population, its importance is up to 0.6. Compared with community home care service stations, the economic level, environment, transportation infrastructure and competition of different types of services are more important to elderly care institutions. This difference occurs mainly because compared with community home care service stations, the distribution of elderly care institutions is not only affected by the distribution of the elderly population, but also needs a beautiful environment and less competition to improve the attractiveness of consumption. A convenient transportation infrastructure is needed to expand the range of services. In addition, elderly care institutions also have higher requirements for the economic capacity and consumption level of the elderly population. Economically developed areas tend to have a higher population consumption capacity, so they are affected by the level of economic development. The importance of the elderly population to community home care stations is much higher than that of elderly care institutions, while the influences of the level of economic development and competition on community home care stations are very weak, with importance levels of only 0.01, which highlights the public property of community home care stations and guarantees the service needs of the elderly population. However, although some streets have elderly populations, the number of community home care service stations is still small. Therefore, it is necessary to increase financial investments and build community public welfare service stations for the elderly to make up for the lack of social capital and market dominance. The number of medical resources is equally important to community home care service stations and elderly care institutions, both of which need the support of medical resources.
Fig. 4 Feature importance of different factors for the distribution of elderly care resources

4 Discussion and conclusions

4.1 Discussion

China’s elderly care service system includes three types of situations: home care, community care and institutional care. According to the “Research Report on Elderly Care Methods and Service Demand in Jiangsu Province” published by the Consumer Protection Commission of Jiangsu Province, 83.3% of the elderly are willing to use home care, 8.8% are willing to use community care, and 7.9% choose institutional care. With the increase of the willingness to use community and institutional care, it is necessary to accelerate the improvement of the elderly care security system and resolve the imbalance between supply and demand, as well as the regional imbalance and the urban-rural imbalance. According to the characteristics of the supply and demand of elderly care services in Wuxi, this paper puts forward the following six suggestions.
First, we should optimize the allocation of elderly care resources and increase investments in elderly care facilities. It is necessary to narrow the regional gap between old-age services and the gap between urban and rural areas. We should actively promote the standardization of elderly care services, establish a sound system of standards for elderly care services, guide and encourage old-age service institutions with a good foundation and strong enthusiasm for carrying out pilot and demonstration work on standardization, and strive to increase the proportion of high-grade old-age care institutions. We should improve the competitiveness of different types of elderly care institutions in terms of price and services, focus on building elderly care institutions for low and middle income groups, moderate the building of elderly care institutions for middle and high income families, and give full play to the role in guaranteeing the bottom line of public elderly care services.
Second, it is necessary to encourage the application of scientific and technological innovation in the field of elderly care. In view of the continuous improvement in the quality characteristics of the elderly population and the high proportion of the young elderly, we should promote the application of Internet of Things technology and implement intelligent elderly care equipment, remote medical services and intelligent health management. This approach would provide efficient, fast, low-cost and intelligent elderly care services, in order to improve the efficiency and quality of elderly care services.
Third, we must promote the fine management of elderly care services. Surveys on the demand for elderly care services should be conducted, and relevant information and data should be collected to provide a basis for formulating policies for the development of elderly care services. Institutions such as guidance centers for elderly care services should be established to regularly carry out work such as quality assessment, safety assessment, and training of elderly care service personnel, which would improve the professional level of elderly care services.
Fourth, it is necessary to build a system of elderly care services that combines medical and nursing care. It is necessary to promote the integration of primary medical and health institutions with home and community elderly care services, support communities in improving the quality and quantity of their elderly care facilities, and increase rehabilitation aids in community elderly care centers. Elderly care institutions should be encouraged and guided to cooperate with surrounding medical and health institutions in accordance with norms, including medical and elderly care institutions that meet the provisions for the scope of medical insurance, and improve the service capacity of integrated medical and elderly care.
Fifth, we should upgrade tourism health service facilities. With the integration of the Yangtze River Delta, the demand for remote tourism and health care may continue to rise. As a city with outstanding tourism resource endowment and location advantages in the Yangtze River Delta region, Wuxi can make use of its high-quality tourism resources, combine tourism with elderly care, and focus on building tourism health care brands in qualified elderly care institutions. It is necessary to cultivate high-grade elderly care institutions near scenic spots and strengthen the supply capacity of tourism health care services such as catering and medical care. We should optimize the tourism health care service procedures, so that the elderly can view the basic information such as the addresses of the institutions, contact information, real scene maps, and spare beds in real time through the smart tourism health care service platform, in order to improve the efficiency and quality of tourism health care services.
Sixth, we should build a cross-domain elderly care service consortium. It is necessary to formulate the development goals of integrated elderly care services in the Yangtze River Delta, strengthen regional cooperation in elderly care services, innovate trans-regional elderly care service mechanisms, promote the trans-regional sharing of elderly care resources, mutual recognition and common standards of elderly care services, open trans-regional elderly care institutions, and promote the docking of elderly care service information with other information resources, such as household registration, medical care, social security, and social assistance. This would allow the elderly to enjoy basic public services and convenient settlement, and promote the construction of an integrated elderly care service system in the Yangtze River Delta.

4.2 Conclusions

In this study, accessibility analysis and the XGBoost algorithm were used to calculate the accessibility of elderly care institutions and community home care service stations in Wuxi City, and the factors influencing the distribution of elderly care institutions and community home care service stations were analyzed. Three main conclusions were reached.
First, the distribution of elderly care institutions in Wuxi City has problems of mismatch between supply and demand and uncoordinated regional development. Regarding the distribution, the elderly care institutions in Liangxi District and Binhu District are the most densely distributed, and most large elderly care institutions are concentrated. In terms of accessibility, Liangxi District, Binhu District and Xishan District have relatively high numbers of accessible beds, while Yixing City and Jiangyin City have relatively low accessibility, and Yixing City and Jiangyin City have the most prominent problems of unbalanced accessibility. From the perspective of high-grade elderly care institutions, Wuxi has three Grade 5 elderly care institutions and 18 Grade 4 elderly care institutions. The number of high-grade elderly care institutions accounts for a relatively low proportion, and all Grade 5 elderly care institutions are located in the Binhu District, so the number of high-grade elderly care institutions is small and their spatial distribution is uneven.
Second, the distribution of community home care service stations is relatively more equitable, but it still does not match the size of the elderly population. The number of home care service stations per thousand people in Dongting Street, which has the richest community home care resources, is 21.98, while it is less than 0.1 in Qingyang Town, Zhutang Town and Lihu Street.
Third, there are obvious differences in the factors affecting the distributions of elderly care institutions and community home nursing service stations. The distribution of elderly care institutions is affected by transportation infrastructure, the size of the elderly population, the number of medical resources, the environment, economic level and other factors; while community home care is mainly affected by the size of the elderly population, highlighting the public property of community home care stations, to guarantee the service needs of the elderly population.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Chen T, Guestrin C. 2016. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. San Francisco, USA: 785-794.

[2]
Dai D. 2010. Black residential segregation, disparities in spatial access to health care facilities, and late-stage breast cancer diagnosis in metropolitan Detroit. Health & Place, 16(5): 1038-1052.

[3]
Ding L, Wang J P, Hu N S. 2022. Fair performance evaluation of resource allocation of urban elderly care institutions under the concept of health equity: A case study of Wuhan City. Modern Urban Research, (8): 1-7. (in Chinese)

[4]
Ding Q X, Zhu L X, Luo J. 2016. Spatial accessibility analysis of elderly care facilities in Wuhan. Human Geography, 31(2): 36-42. (in Chinese)

[5]
Gao X L. 2013. The characteristics of urban residents’ preference for old-age care institutions and community differences. China Soft Science, (1): 103-114. (in Chinese)

[6]
Han F, Luo R C. 2020. Research on supply and demand matching of home care services in urban community based on accessibility measurement: A case study of Nanjing. Economic Geography, 40(9): 91-101. (in Chinese)

DOI

[7]
Handy S L, Niemeier D A. 1997. Measuring accessibility: An exploration of issues and alternatives. Environment and Planning A, 29(7): 1175-1194.

DOI

[8]
Jiang L, Chen X Y, Zhu H. 2021. Spatial distribution characteristics and causes of differentiation of urban nursing homes in China. Acta Geographica Sinica, 76(8): 1951-1964. (in Chinese)

[9]
Li Z G, Yu Y J, Jiang S Y, et al. 2023. Research on the allocation mode of affordable elderly care facilities in large cities: A case study of the main urban area of Wuhan. Planners, 39(1): 26-34. (in Chinese)

[10]
Radke J, Mu L. 2000. Spatial decompositions, modeling and mapping service regions to predict access to social programs. Geographic Information Sciences, 6(2): 105-112.

[11]
Tao Z L, Cheng Y. 2016. Progress in two-step mobile search method and its extended forms. Progress in Geography, 35(5): 589-599. (in Chinese)

[12]
Wang F. 2012. Measurement, optimization, and impact of health care accessibility: A methodological review. Annals of the Association of American Geographers, 102(5): 1104-1112.

PMID

[13]
Wang Z G, Lin R, Yang Y M, et al. 2023. Research on spatial distribution and accessibility of institutional elderly care facilities in Chengdu. Science of Surveying and Mapping, 48(1): 236-244, 276. (in Chinese)

[14]
Xu K, Tu P. 2020. Evaluation on spatial accessibility of elderly care service facilities in Fuzhou. Journal of Huaqiao University (Natural Science), 41(3): 340-347. (in Chinese)

[15]
Xu X, Zhao Y. 2017. Spatial distribution pattern and accessibility evaluation of elderly care service facilities in Nanjing: A two-step mobile search method based on time cost. Modern Urban Research, (2): 2-11. (in Chinese)

[16]
Yin W H, Liu S. 2021. Study on spatial distribution characteristics and accessibility of elderly care institutions in Shanghai. Modern Urban Research, 36(6): 17-23. (in Chinese)

[17]
Zhang H Y, Zhang B R. 2016. Research on spatial layout of home care facilities in urban communities: A case study of central urban area of Shanghai. Yunnan Geographical Environment Research, 28(1): 55-59. (in Chinese)

[18]
Zhang L L, Bi S B, Liu S L, et al. 2023. Research on the accessibility and supply-demand relationship of elderly care institutions based on traffic data: A case study of Beijing. Geography and Geo-Information Science, 39(2): 81-88. (in Chinese)

[19]
Zhao D X, Han Z L, Wang Li, et al. 2014. Research on the accessibility of urban home care service facilities based on two-step mobile search method: A case study of the younger elderly in Shahekou District of Dalian City. Areal Research and Development, 33(6): 27-32. (in Chinese)

Options
Outlines

/

〈 〉