Spatiotemporal Evolution and Influencing Factors of Population Aging in the Triangle of Central China at Multiple Scales

Abstract

This study focuses on the Triangle of Central China and investigates the spatiotemporal evolution, driving factors, and impacts of population aging on regional sustainable development from 2000 to 2020. The study adopts an innovative two-scale analytical framework at the prefecture and district/county level, integrating spatial autocorrelation analysis, the Geodetector model, and geographically weighted regression. The results show a significant acceleration in population aging across the study area, accompanied by pronounced spatial clustering, particularly in western Hubei and the Wuhan metropolitan area. Over time, the spatial distribution has evolved from a relatively dispersed pattern to one of high concentration. Key drivers of the spatial heterogeneity of aging include economic disparities, demographic transitions, and the uneven spatial allocation of public services such as healthcare and education. These aging patterns profoundly affect the region’s potential for sustainable development. Accordingly, the study proposes a multi-scale collaborative governance strategy: At the prefecture level, efforts should focus on promoting the coordinated development of the silver economy and optimizing the spatial redistribution of healthcare resources; At the district and county level, priorities should include strengthening infrastructure, curbing the outflow of young labor, and improving access to basic public services. By integrating spatial analysis techniques with sustainable development policy recommendations, this study provides a basis for scientifically measuring, understanding, and managing demographic transitions. This is essential for achieving long-term socioeconomic sustainability in rapidly aging regions.

1. Introduction

Population aging typically refers to the proportion of people aged 65 and above exceeding 7% of the total population, a key indicator of a society entering the aging stage. Since Raymond Pearl first introduced the concept of “population aging” in 1940, the global trend of aging has intensified [1]. In 1950, only nine countries had more than 10% of their population aged 65 or above [2]. According to United Nations data, people aged 65 and above accounted for 10% of the global population in 2022, and this is projected to increase to 16% by 2050, outpacing all other age groups. Developing countries, in particular, face the acute challenge of “growing old before becoming rich” [3]. Since entering an aging society in 1999, China has exhibited characteristics such as a large elderly population, rapid growth, and uneven regional development [4,5]. By the end of 2022, China’s population aged 65 and over reached 210 million, accounting for 14.9% of the total population [6]. It is projected that by 2035, this proportion will exceed 21%, marking China’s transition into a “super-aged society” [7]. This transition is accompanied by a series of sociological phenomena such as “aging before affluence,” “unprepared aging,” and “aging alone” [8,9,10]. Consequently, the contradiction between aging and economic development will intensify, posing a strategic challenge that China must confront.

Academic research on aging has primarily focused on regional disparities [11,12], development trends [13,14], spatial distribution [15,16], and its impact on economic development [17,18]. Previous studies have mostly been conducted at the provincial, prefecture, and county levels, along with cross-national or interregional comparisons. These studies reveal spatial differentiation patterns of population aging, including disparities between eastern, central, and western regions; coastal and inland areas; urban cores and peripheries; concentric structures; core-periphery structures; and urban-rural differences.

At the international level, Studies by Cowgill [19] and Golant [20] indicate that elderly populations in the United States are concentrated in urban cores and remote rural areas, forming so-called “retirement centers.” Heikkila [21] noted significant regional differences in the distribution of population aging across provinces, cities, and various regions in Finland. Novee Lor Leyso [13] used grid-based analysis to reveal the evolution of aging types in Tokyo from 2000 to 2015. Rogers [22] and others compared the spatial distribution patterns of elderly populations in four countries, including the US and UK, finding concentrations in urban cores and remote rural areas, along with a trend of migration toward suburban areas. Lee Ji Hey and Kim Hyoung Joong [23] focused on the Seoul metropolitan area to identify and analyze the spatial distribution and clustering characteristics of the elderly population. Zhu Wei and Yuan Chao [24] analyzed the spatial distribution of the aging population in Singapore to address urban heat-related health risks. González A.L. and Vázquez J.A.A. [25] examined the impact of immigration on population aging in Spain. Andreas [26] synthesized the analyses of numerous scholars and explained the underlying processes behind the rapid population aging in Central and Eastern Europe. In addition, the study discusses the main demographic drivers behind this trend in each country examined.

In domestic research, Wu Linguo [14], based on the Hu Huanyong Line demarcation, the regional disparities in aging among China’s prefecture-level administrative divisions narrowed progressively between 2000 and 2020. Li Shuai et al. [27] discovered a spatial clustering of population aging in the Beijing–Tianjin–Hebei region, accompanied by an urban-rural “inverted” pattern. In addition, population aging in various provinces exhibits diverse spatial distribution patterns: Jiangsu Province forms an “E-shaped” structure [28]; Shandong Province shows a stepped distribution [29]; Sichuan Province is undergoing a transition to deep aging [30]; Guizhou Province shows a gradient decrease from northeast to southwest [31]; and Chongqing presents a core-periphery sandwich structure between its main city and suburban areas [32]. These studies systematically depicted the spatiotemporal evolution of population aging using geographic information technology and statistical modeling methods, offering valuable references for policy-making.

The influencing factors of population aging can be broadly categorized into two aspects, demographic factors and socioeconomic factors. Demographic factors include both natural and mechanical population changes. First, regarding natural changes, declining mortality rates and increasing life expectancy are the primary driving forces. Meanwhile, a decrease in infant mortality has increased the proportion of the working-age population, partially offsetting the effects of aging [33,34]. For example, in Fujian, the growth rate of the working-age population exceeds that of the elderly group [35]. Meng Linguo [36] pointed out that although the universal two-child policy has had a mitigating effect to some extent, its overall impact remains limited. Second, in terms of population migration, some scholars argue that migration alters the age structure. Cities attract young people, alleviating urban aging while intensifying aging in rural areas [37,38,39]. Studies by Zhang Hangkong [40] and others indicate that interprovincial migration influences population aging in China—raising aging levels in some provinces and lowering them in others—thus narrowing the aging gap between provinces. The development of urban agglomerations promotes population concentration in core cities. In some migrant destinations such as Shenzhen, the aging level has remained low for a long time [34,41,42,43]. In addition, the “migratory bird-style” elderly care phenomenon increases the complexity of aging patterns in economically developed regions [44,45,46].

Economic factors also exhibit regional disparities. Zhou Peng [30,47] pointed out that per capita GDP has a greater impact in plain areas, while its effect is less significant in hilly and mountainous regions. Thaveesha Jayawardhana [48] and colleagues examined the relationship between the elderly population and economic growth across 25 countries in the Americas. Their findings indicate a unidirectional causal relationship in some of these countries. Moreover, economic growth exerts a positive influence on the aging population in North America, while in South America, the effect is negative. Other findings suggest that in the early stage of urbanization, the “siphon effect” improved urban population structures but exacerbated rural aging. The specific impact mechanisms in shrinking cities during the mid-to-late stages require further investigation [49,50]. Educational attainment is positively correlated with aging rates by reducing fertility intentions [39,42]. Medical advancements show greater sensitivity in underdeveloped areas due to favorable policy bias [51].

As a key economic growth pole in China, the Triangle of Central China has faced an increasingly prominent problem of population aging in recent years, posing a major challenge to its sustainable development. Between 2000 and 2020, the aging rate in this region rose from 6.72% to 13.83%. The population aged 65 and above reached 17.48 million, accounting for 9.17% of the national elderly population. The region has entered a stage of moderate aging and is rapidly approaching a deeply aged society. This trend is not only a natural outcome of demographic shifts but also the result of multiple factors, including economic transformation and imbalances in public services. If not properly addressed, the aging problem will severely impact regional economic vitality, the social welfare system, and the provision of public services, thereby undermining the region’s position in China’s economic landscape. Therefore, how to actively address aging has become a pressing issue for the Triangle of Central China.

Although some scholars have revealed the spatial and temporal patterns of population aging in the Yangtze River Basin, current research still has certain limitations. Based on this, the innovations and contributions of this study are reflected in the following aspects: First, this study fills a gap in existing literature by moving beyond the previous focus on the upper Yangtze region and the Yangtze River Delta. It is the first to systematically analyze the spatial and temporal distribution of aging and its influencing factors in the Triangle of Central China. Second, this study adopts an innovative approach by combining prefecture-level and district and county-level scales. This breaks through the limitations of traditional single-scale analysis and reveals differences in influencing factors across scales, offering a new perspective for more refined research. Third, in terms of methodology, this study integrates the Geodetector model with the Geographically Weighted Regression (GWR) model to construct a spatial analysis framework that links global and local scales. This framework quantifies the statistical significance and explanatory power of each factor while capturing spatial non-stationarity and regional differences, thus offering a new technical pathway for analyzing the spatial determinants of aging.

In summary, the population aging issue in the Triangle of Central China not only affects the region’s high-quality socioeconomic development but is also closely linked to the national strategies for addressing aging. This study selects 31 prefecture-level cities and 173 counties in the Triangle of Central China as research subjects, analyzing the characteristics of population aging from 2000 to 2020. Using geographic visualization models, spatial autocorrelation indices, and the geographical detector method, this study reveals the spatial distribution patterns and influencing mechanisms of aging, providing a basis for policy formulation and resource optimization.

2. Data Sources and Research Methodology

2.1. Overview of the Study Area

The city cluster in the middle reaches of the Yangtze River is a nationally designated urban agglomeration centered around the Wuhan Metropolitan Area, Changsha–Zhuzhou–Xiangtan City Group, and Poyang Lake City Group as the core (Figure 1). In 2015, the National Development and Reform Commission designated the region as a new economic growth pole, a demonstration zone for new urbanization, a pilot area for inland openness and cooperation, and a model for building a resource-efficient and environmentally friendly society. In 2018, the central government reaffirmed Wuhan’s role as the core city driving regional development and underscored its strategic importance in both the Yangtze River Economic Belt and the Rise of Central China strategy.

The study area is located in the central Yangtze River Basin, extending from Chongqing to Shanghai, with Wuhan serving as the primary transportation and economic hub. Changsha–Xiangtan serves as the western growth pole, while Nanchang anchors the eastern part of the region. The region features diverse topography: the industrialized Jianghan Plain in the north; ecological buffer zones formed by hills and mountains in central Hunan and northern Jiangxi; and low hills and plains in eastern Hubei that support industrial expansion. The study area includes 13 prefecture-level cities and 59 districts and counties in Hubei Province, 8 prefecture-level cities and 46 districts and counties in Hunan Province, and 10 prefecture-level cities and 68 districts and counties in Jiangxi Province.

2.2. Data Sources

The core data required for this study include geographic information, demographic statistics, and socioeconomic data. This study conducted spatial analysis based on the ArcGIS 10.8 platform.

(1)

Vector data of regional administrative boundaries. The data were obtained from the National Administrative Division Information Query Platform of the People’s Republic of China (http://xzqh.mca.gov.cn/map/, accessed on 27 June 2020), using the 2020 administrative divisions. To address frequent adjustments at the district and county levels, the 2000 and 2010 statistical and geographic data were merged based on the 2020 administrative boundaries. This approach ensures data consistency and reliability and facilitates comparison across datasets. In this study, to present city names more clearly on the map, we simplified the names of prefecture-level cities and autonomous prefectures. For example, “Wuhan City” was simplified to “Wuhan,” “Nanchang County” to “Nanchang,” and “Changyang Tujia Autonomous County” to “Changyang Tujia.” This simplification helps to highlight key information on the map and allows readers to more intuitively grasp the spatial distribution of cities.

(2)

Demographic data. The data were sourced from the 5th (2000), 6th (2010), and 7th (2020) national population censuses published by the National Bureau of Statistics. Data were primarily collected for 31 prefecture-level cities and 173 districts and counties in the Triangle of Central China, covering both the permanent population and the population aged 65 and over.

(3)

Socioeconomic data. The main sources include the 2001, 2011, and 2021 statistical yearbooks of Hunan, Hubei, and Jiangxi provinces, as well as those of the 31 prefecture-level cities. To obtain a more comprehensive picture of regional economic and social development, the study also referred to the China Regional Economic Statistical Yearbook, China County Statistical Yearbook, and China City Statistical Yearbook.

(4)

Selection of Influencing Factor Indicators

Theoretically, the simultaneous decline in the young population and the increase in the elderly population were direct drivers of population aging. However, under the backdrop of rapid urban-rural integration, regional disparities in economic development and education also influenced the progression of population aging, adding complexity to its driving mechanisms. According to relevant studies [52], population aging was driven by both demographic structural changes and external socioeconomic factors. Therefore, this study selected nine indicators across demographic, social, and economic dimensions to examine the differential mechanisms of population aging in the Triangle of Central China (

Table 1. Indicators and Selection Rationale.

	Variable	Indicator Description	Calculation Method	Data Source
Demographic Factors	Proportion of Population Aged 55–65	Trend of Population Aging	Ratio of Population Aged 55–65 in Previous Period	National Census
	Natural Population Growth Rate	Conditions for a Younger Population Structure	Birth Rate minus Death Rate	National Census
	Net Migration Rate	Regional Population Mobility	(In-Migrants—Out-Migrants)/Resident Population	National Census
Economic Factors	Per Capita GDP	Economic Push and Pull Factors of Migration	Regional GDP/Resident Population	Statistical Yearbook
	Value Added by Primary Industry	Final Output of Agricultural Activities	Total value added of agriculture, forestry, animal husbandry and fishery	Statistical Yearbook
	Proportion of Secondary and Tertiary Industries	Regional Economic Structure	(Secondary + Tertiary Industries)/Regional GDP	Statistical Yearbook
Social Factors	Urbanization Rate	Level of Urban Development	Urban Population/Total Population	National Census
	Hospital Beds per 1000 People	Objective Factors for Longer Life Expectancy	Total Hospital Beds/Resident Population × 1000	Statistical Yearbook
	Illiteracy Rate	Basic Education Level	(Number of illiterate people aged 15 and above ÷ Number of people aged 15 and above) × 100%	National Census

).

2.3. Research Methods

(1)

Definition of Population Aging

This paper defines the degree of population aging based on the aging rate, which refers to the proportion of people aged 65 and above in the total resident population. This proportion is referred to as the aging coefficient. According to United Nations standards, the degree of population aging is classified as follows: areas where the elderly (aged 65 and above) account for less than 7% of the total population are considered non-aging societies; 7% to 10% as mildly aging; 10% to 14% as moderately aging; 14% to 20% as severely aging; and more than 20% as super-aged societies.

(2)

Spatial Autocorrelation

Spatial autocorrelation identifies the spatial relationships of geographic phenomena. This study uses the Global Moran’s I index to assess whether population aging in the Triangle of Central China exhibits spatial clustering. The formula is as follows:

$$\begin{array}{c}I={\displaystyle \frac{n\sum _{i=1}^n\sum _{j=1}^n{w}_{ij}\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}{\left(\sum _{i=1}^n\sum _{j=1}^n{w}_{ij}\right)\sum _{i=1}^n{\left(x_i-\overline{x}\right)}^2}}\end{array}$$

(1)

In the equation, $I$ represents the global autocorrelation index; $n$ is the number of prefecture-level or district units; $x_i$ is the aging rate of the $\mathrm{i}$ district unit; $\overline{x}$ is the average aging rate of the prefecture or district level; $w_{ij}$ is the spatial weight matrix, defined such that the weight decreases with distance. The Moran’s I index has a value range of [−1, 1]. An index value of 1 indicates a high positive spatial correlation, −1 indicates a high negative spatial correlation, and 0 indicates no spatial correlation.

The Local Indicator of Spatial Association (LISA) further identifies the dominant types of regional clustering. The calculation formula is as follows:

$$\begin{array}{c}{I}_i={\displaystyle \frac{n\left(x_i-\overline{x}\right)}{\sum {\left(x_i-\overline{x}\right)}^2}}n{\displaystyle {\displaystyle \sum }_{j=1}^n}{w}_{ij}\left(x_j-\overline{x}\right)\end{array}$$

(2)

In this formula, $I_i$ is the local spatial autocorrelation index. The results are categorized into four types: High-High (H-H), High-Low (H-L), Low-High (L-H), and Low-Low (L-L).

(3)

Geographic detector

This study utilizes the factor detection module of the Geodetector model to analyze the influencing factors and interactive effects contributing to the spatial differentiation of population aging in the Triangle of Central China. Factor detection quantifies how much each influencing factor explains the spatial heterogeneity of population aging. The calculation formula is as follows:

$$\begin{array}{c}q=1-{\displaystyle \frac{\sum _{h=1}^L{N}_h{M}_h^2}{N{M}^2}}=1-{\displaystyle \frac{SSW}{SST}}\end{array}$$

(3)

$$\begin{array}{c}SSW={\displaystyle {\displaystyle \sum }_{h=1}^L}{N}_h{M}_h^2,SST=N{M}^2\end{array}$$

(4)

In the equation, $q$ represents the explanatory power of each individual factor on the spatial distribution of aging. $h\in [0,9]$ denotes the stratification of the aging rate and its influencing factors. As can be seen in Table 1 below, this paper selects 9 influencing factor indicators and uses the “Jenk optimal” natural breakpoint method to divide the population aging rate and influencing factors into 9 categories, so the $h$ layer is 9. $N_h$ and $N$ represent the number of units in stratification $h$ and in the entire dataset, while $M_h^2$ and $M^2$ represent the variances of the aging rate in stratification h and for the whole dataset, respectively. $SSW$ and $SST$ represent the sum of within-class variance and the total variance of the entire dataset, respectively, $q\in [0,1]$. The larger the $q$ value, the stronger the explanatory power of the influencing factors on the dependent variable; conversely, a smaller $q$ value indicates weaker explanatory power.

(4)

Geographically Weighted Regression

Traditional linear regression models provide only “average” or “global” estimates. When explanatory variables exhibit spatial characteristics and spatial autocorrelation, the assumption of independence in residuals, as required by Ordinary Least Squares (OLS), is violated. The Geographically Weighted Regression (GWR) model accounts for spatial non-stationarity by allowing parameter estimates to vary across geographic locations, thereby capturing local variations in relationships and improving the realism of the results. However, the GWR model uses a common bandwidth average across all variables, which may introduce estimation bias. The calculation formula is as follows:

$$\begin{array}{c}{y}_i=b_o\left(p_i,q_i\right)+{\displaystyle {\displaystyle \sum }_{t=1}^p}{b}_t\left(p_i,q_i\right)\times x_{it}+{\epsilon }_i\end{array}$$

(5)

where $(p_i,q_i)$ represents the geographical coordinates of the $i$-th sampling point, $b_t(p_{i }, q_i)$ is the $t$-th regression parameter at the $i$-th sampling point, $b_0(p_i, q_i)$ is the regression constant at the $i$-th sampling point, ${\epsilon }_i$ represents the error term, where, ${\epsilon }_i~N\left(0, \sigma \right), Cov({\epsilon }_i,{\epsilon }_j)=0,i\ne j$.

3. Spatiotemporal Characteristics of Population Aging in the Triangle of Central China

3.1. Analysis of the Spatiotemporal Evolution of Aging Levels in the Triangle of Central China

3.1.1. Prefecture-Level Scale

(1)

Analysis of Aging Levels

Aging coefficient data from 31 prefecture-level cities in 2000, 2010, and 2020 were used, and scatter plots were created at 10-year intervals. The x-axis represented the cities within the Triangle of Central China, and the y-axis represented the aging coefficient for each corresponding year (Figure 2). The aging coefficients increased across all three time points, with a particularly sharp rise between 2010 and 2020.

(2)

Spatiotemporal Distribution of Aging Levels

To further visualize the spatial variation in aging levels within the Triangle of Central China, ArcGIS was used to map aging categories at the prefecture-level scale (Figure 3). The results showed that in 2000, the region was predominantly characterized by mild aging, with affected cities clustered in the western part of the Triangle, forming a contiguous distribution. By 2010, aging had intensified significantly. Only six prefecture-level cities had reached moderate aging levels, indicating the gradual expansion of the aging issue across the region. By 2020, all cities in the region had aging rates exceeding 14.00%. Most prefecture-level cities in Hubei and Hunan experienced a deep aging, while those in Jiangxi were mainly at a moderate level. Spatially, the aging pattern was characterized by lower levels in the east and higher levels in the west. There were notable differences between the northern and southern parts of the Yangtze River: the north and west experienced deep aging, while the south had moderate aging. Overall, the population aging in the region accelerated, showing a spatially connected, clustered development pattern. Deep aging has become an inevitable trend.

3.1.2. District and County Level

(1)

Analysis of Aging Levels

The degree of population aging in the Triangle of Central China has intensified rapidly, as shown in

Table 2. Proportions of Population Aging Types at District and County Level in the Triangle of Central China.

Types of Population Aging	Non-Aging (0–7%)		Mild Aging (7–10%)		Moderate Aging (10–14%)		Severe Aging (14–20%)		Aged Society (>20%)
Year	Num.	Per.	Num.	Per.	Num.	Per.	Num.	Per.	Num.	Per.
2000	111	64.53	60	34.88	1	0.58	0	0	0	0
2010	22	12.79	102	59.30	43	25	3	1.74	2	1.16
2020	1	0.58	2	1.16	80	46.51	79	45.93	10	5.81

. In 2000, the region was mainly characterized by non-aging and mildly aging counties, with non-aging areas accounting for the majority (64.53%), and only one county classified as moderately aging. By 2010, the dominant categories shifted to mild and moderate aging (together making up 84.30%). Additionally, three counties were newly classified as severely aging and two as aging societies. By 2020, moderate and severe aging became predominant (92.44% combined), with the number of aging society counties increasing to ten. Overall, from 2000 to 2010, the aging rate increased more rapidly—nearly doubling—mainly due to the significant rise in mildly aging counties. From 2010 to 2020, severely aging counties experienced the largest increase, reaching 45.93%.

(2)

Spatiotemporal Distribution of Aging Levels

Spatially, the variation in aging levels at the county scale was more complex than at the prefecture-level scale, with a wider range of aging types represented (Figure 4). In 2000, a spatial pattern of “more severe in the west, milder in the east” emerged. Areas with higher aging levels (moderate or above) were primarily concentrated in western Hubei (Wufeng, Hefeng), southwestern Hunan (Xinhuang, Tongdao), and southwestern Jiangxi (Ningdu). Core cities such as Wuhan, Changsha, and Nanchang, along with their surrounding counties, remained in non-aging or mildly aging stages, indicating relatively youthful populations. By 2010, regional disparities had widened, and the “west-deep, east-shallow” spatial trend further intensified. Some counties like Xiangyang, Jingzhou, Jiujiang, and Jingdezhen still remained non-aging, while deep aging appeared in northwest Hubei. By 2020, severe aging was concentrated in counties such as western Hubei, western and central Hunan, and northern Jiangxi. Although aging also increased in core urban areas like Changsha, Wuhan, and Nanchang, these regions still appeared less aged overall. Overall, from 2000 to 2020, the spatial distribution of population aging in the Triangle of Central China shifted from scattered to widespread clusters, displaying a dynamic pattern of diffusion. The overall trend showed heavier aging in the northwest compared to the southeast, divided by the Yangtze River.

3.2. Spatial Differentiation and Temporal Variation of the Elderly Population in the Triangle of Central China

3.2.1. Global Spatial Clustering Characteristics of the Aging Rate

(1)

Prefecture-level Scale

The Moran’s I index of the aging rate in the Triangle of Central China was 0.521 in 2000, 0.165 in 2010, and 0.527 in 2020, indicating strong positive spatial autocorrelation in population aging during these years (

Table 3. Global Moran’s I for the Triangle of Central China (2000, 2010, and 2020).

Year	Prefecture-Level Scale			District and County Scales
	Moran’s i	p-Value	Z-Score	Moran’s i	p-Value	Z-Score
2000	0.52096	0.0001	4.909	0.602	0	10.99
2010	0.16529	0.0722	1.797	0.299	0	5.73
2020	0.52712	0.0001	4.929	0.604	0	11.11

). From 2000 to 2010, the Moran’s I index declined. This trend could be attributed to regional policies such as improved transportation infrastructure, industrial restructuring, and talent initiatives, which enhanced interregional population mobility and alleviated local aging pressures. However, the Moran’s I index increased again from 2010 to 2020, suggesting a resurgence in spatial clustering. The primary reason lay in the continued regional imbalance. Compared with the developed eastern regions, the Triangle of Central China faced disadvantages in employment opportunities and income levels, resulting in the outmigration of working-age populations and exacerbating the aging issue. The disparity in public services and infrastructure between urban and rural areas, particularly the lack of healthcare and elderly care resources in rural regions, further intensified rural aging.

(2)

District and county level Scale

At the district and county level, the Moran’s I values for the aging rate were 0.602 in 2000, 0.299 in 2010, and 0.604 in 2020, following a trend similar to the prefecture-level analysis: strong clustering that weakened and then intensified. However, the county-level Moran’s I values were consistently higher than those at the prefecture level, indicating stronger local clustering in smaller spatial units. Moreover, the p-values in all three years were 0, and the Z-values were significantly higher than those at the prefecture level, demonstrating that the spatial pattern of aging at the district and county level was more stable and statistically significant.

3.2.2. Local Spatial Clustering Characteristics of the Aging Rate

While global spatial autocorrelation revealed the overall spatial distribution trend of population aging in the Triangle of Central China, it failed to accurately reflect spatial heterogeneity at the local level. Therefore, this study employed the GeoDa tool to calculate and analyze local spatial autocorrelation at both the prefecture level and district/county scales, revealing the spatial clustering patterns and distribution characteristics of the aging population within the region (Figure 5).

(1)

Prefecture-level Scale (Figure 5A)

In 2000, the region exhibited clear polarization: H-H clustering zones with high aging rates were mainly concentrated in six cities, including Changsha and Hengyang in Hunan, reflecting more severe aging in the southwest. L-L clustering zones were found in Ezhou and Huangshi in Hubei and Jiujiang and Yingtan in Jiangxi, forming regions dominated by younger populations. Wuhan presented an H-L outlier pattern, with a highly aging city surrounded by low-aging regions. By 2010, the H-H clusters had shrunk to Yiyang and Loudi in Hunan, while L-L clusters had shifted to the Jiujiang-Yingtan river belt in Jiangxi, forming a contiguous low-aging region in the east. By 2020, H-H clusters had expanded northward to prefecture-level cities including Jingmen, Jingzhou, Tianmen, Qianjiang, and Xiantao in Hubei, while L-L zones had further expanded to include Xinyu, Fuzhou, and Yichun in Jiangxi. Overall, H-H zones had gradually shifted from southern Hunan and Jiangxi to northern Hubei, while L-L zones had moved eastward.

(2)

District and county level Scale (Figure 5B)

In 2000, H-H clustering zones were mainly concentrated in certain counties in western Hubei and central Hunan. L-L clusters were located in eastern Hubei and eastern Jiangxi. Tuanfeng County in Hubei represented a typical H-L outlier, with a higher aging level than the surrounding areas. By 2010, H-H clusters had significantly contracted, remaining only in counties such as Shimen, Songzi, Yidu, and Zhijiang. L-L clusters had shifted to Jiangxi Province, forming a new low-aging zone. For the first time, L-H clusters had emerged, represented by Xiangyang City in Hubei, indicating uneven aging across the region. In 2020, H-H clustering zones had expanded once again, forming new high-aging clusters in areas such as Jingzhou and Qianjiang in Hubei and Changde in Hunan. Meanwhile, L-L zones had continued to grow, extending into additional counties in Jiangxi. Notably, Jinxian County in Jiangxi exhibited an H-L clustering pattern, whereas Jingmen in Hubei and Loudi in Hunan showed L-H characteristics.

In summary, from 2000 to 2020, the Triangle of Central China showed an increasing trend of spatial clustering in population aging. High-aging zones had shifted northward, while low-aging areas had mainly concentrated in Jiangxi with relatively continuous distribution. From a spatial scale perspective, the prefecture level was suitable for grasping overall aging trends and regional planning, while the district and county levels were more effective in identifying local differences and marginal characteristics, which was crucial for optimizing elderly care facilities and policy targeting.

4. Analysis of Influencing Factors

This study utilizes ArcGIS 10.8 as the analytical platform, integrating the Geographical Detector Model with Geographically Weighted Regression (GWR) to establish a spatial explanatory framework that connects global explanatory power with local spatial heterogeneity. First, the Geographical Detector was applied to examine variations in the quantitative indicators of population aging’s influencing factors, identifying several key drivers. Subsequently, the GWR model was employed to further explore the spatial heterogeneity in the impact of these factors across different regions. Finally, policy recommendations were formulated based on the research findings.

4.1. Detection Results of Univariate Influencing Factors of Population Aging

This study employs ArcGIS 10.8 to conduct hierarchical clustering of the Triangle of Central China at both the prefecture level and district and county levels. The Geodetector model is then used to evaluate the explanatory power and significance of various factors influencing the aging population rate across spatial and temporal dimensions (

Table 4. Determining the Power of Influencing Factors on Population Aging at the Prefecture Level in the Triangle of Central China.

Variant	2000 Year		2010 Year		2020 Year
	q Statistic	p Value	q Statistic	p Value	q Statistic	p Value
Natural Population Growth Rate	0.342	0.000	0.507	0.169	0.405	0.027
Proportion of Population Aged 55–65	0.630	0.000	0.601	0.000	0.535	0.001
Net Migration Rate	0.091	0.000	0.270	0.000	0.233	0.001
Per Capita GDP	0.242	0.000	0.078	0.000	0.409	0.001
Value Added by the Primary Industry	0.586	0.000	0.444	0.000	0.537	0.000
Proportion of Secondary and Tertiary Industries	0.266	0.001	0.586	0.000	0.452	0.085
Hospital Beds per 1000 People	0.162	0.000	0.155	0.000	0.366	0.000
Illiteracy Rate	0.547	0.149	0.291	0.000	0.351	0.000
Urbanization Rate	0.271	0.000	0.303	0.000	0.366	0.000

and

Table 5. Determining the Power of Influencing Factors on Population Aging at the District and County Level in the Triangle of Central China.

Variant	2000 Year		2010 Year		2020 Year
	q Statistic	p Value	q Statistic	p Value	q Statistic	p Value
Natural Population Growth Rate	0.437	0.000	0.232	0.000	0.347	0.000
Proportion of Population Aged 55–65	0.305	0.000	0.404	0.002	0.269	0.000
Net Migration Rate	0.100	0.000	0.214	0.000	0.248	0.000
Per Capita GDP	0.217	0.000	0.235	0.000	0.236	0.000
Value Added by the Primary Industry	0.405	0.000	0.624	0.003	0.532	0.000
Proportion of Secondary and Tertiary Industries	0.209	0.000	0.528	0.000	0.378	0.000
Hospital Beds per 1000 People	0.105	0.000	0.268	0.000	0.287	0.000
Illiteracy Rate	0.272	0.000	0.078	0.000	0.090	0.000
Urbanization Rate	0.120	0.000	0.090	0.000	0.258	0.000

). The results indicated that under a multi-scale analytical framework, the driving mechanisms of population aging exhibited significant spatial heterogeneity and temporal variation.

(1)

Comparative Influence of Demographic Factors

Among demographic variables, the natural population growth rate showed consistent and significant explanatory power at both spatial scales. Overall, its explanatory power was higher at the district and county level (e.g., q = 0.437 in 2000 and q = 0.347 in 2020) than at the prefecture level (q = 0.342 and 0.405, respectively), indicating that variations in fertility rates and healthcare conditions were more prominent in smaller spatial units. The proportion of the 55–65 age group, an indicator of the accumulation of aging in earlier stages, demonstrated stronger explanatory power at the prefecture level (q = 0.630 in 2000, q = 0.535 in 2020) than at the district and county level (q = 0.305 and 0.269, respectively), reflecting long-term trends in the urban population structure. The net migration rate exhibited slightly higher explanatory power at the district and county level (e.g., q = 0.248 in 2020) than at the prefecture level (q = 0.233), suggesting that grassroots regions responded more directly to population mobility, especially the outflow of younger people and concentration of elderly populations.

(2)

Impact of Economic Structure on Population Aging

At both the prefecture level and district and county levels, the added value of the primary industry served as a key explanatory factor for the spatial differences in population aging. For example, the q-values at the prefecture level were 0.586 in 2000 and 0.537 in 2020, while at the district and county level, they reached 0.624 in 2010. These figures highlighted that the continued outflow of labor from agriculture-dominated regions was a major driver of spatial differentiation in population aging. The explanatory power of per capita GDP was more pronounced at the prefecture level (q = 0.409 in 2020), while at the district and county level it was relatively weak (q = 0.236). The proportion of the secondary and tertiary industries was significant at both spatial levels, but more sensitive at the county scale (q = 0.528 in 2010 and still significant at 0.378 in 2020), whereas the prefecture-level q-value of 0.452 in 2020 was not statistically significant (p = 0.085). This indicated that industrial structure continued to shape population distribution and aging patterns at the district and county levels.

(3)

Comparative Analysis of the Impact of Social Resources and Urbanization Levels

In terms of social services, the number of hospital beds per capita, as an indicator of healthcare resources, demonstrated explanatory power at both the prefecture level (q = 0.366 in 2020) and district and county levels (q = 0.287), with a stronger effect at the prefecture level. The spatial explanatory power of the illiteracy rate also varied significantly by scale, with a higher value at the prefecture level (q = 0.351) than at the district and county level (q = 0.090). The urbanization rate also showed greater explanatory power at the prefecture level (q = 0.366 in 2020) compared to the district and county level (q = 0.258). These findings indicated that the spatial distribution of social resources not only affected inter-city population structures but also revealed emerging intra-city imbalances, particularly in access to healthcare and educational services.

In summary, scale effects significantly influenced the identification of spatial driving mechanisms behind population aging. The prefecture-level scale was better suited to identifying macro-level development disparities, urban cluster structures, and broad resource allocation effects, while the district and county levels were more effective at uncovering micro-level mechanisms such as population migration, uneven local services, and industrial disparities. Therefore, when developing regional aging response strategies, it was essential to consider scale differences and adopt a combined policy approach of “prefecture-level coordination and county-level implementation.” Prefecture-level efforts focused on equitable resource distribution, controlling the influx of elderly populations, and planning for eldercare infrastructure. County-level efforts aimed to improve basic medical and educational resource coverage, optimize urban–rural industrial layouts, and attract and retain younger populations to mitigate the spillover effects of aging.

4.2. Geographically Weighted Regression Analysis of Comprehensive Influencing Factors on Population Aging (Spatial Heterogeneity Analysis–GWR)

4.2.1. Natural Population Growth Rate and Population Aging

(1)

At the prefecture level (Figure 6A)

In 2000, cities like Wuhan, Yichang, and Changde showed strong positive correlations between natural population growth and aging, due to limited outmigration, conservative fertility views, and high birth rates. In contrast, Yueyang, Xianning, and Huangshi showed strong negative correlations, driven by urbanization, delayed marriage, and rising child-rearing costs.

By 2010, positive correlations declined, and cities like Wuhan and Jingdezhen shifted to negative correlations, reflecting industrialization and service sector growth, which increased housing and education pressures and reduced fertility. Meanwhile, some cities such as Changde and Yiyang maintained positive correlations.

In 2020, cities like Yichang, Changde, and Xiangtan once again stood out for strong positive correlation, while provincial capitals Wuhan, Changsha, and Nanchang shifted to strong negative correlation.

(2)

At the district and county level (Figure 6B)

In 2000, strong positive correlations were widespread, especially in central and southern Hunan and western Hubei. These areas were economically underdeveloped, had poor transport, low youth outmigration, and strong cultural preferences for multiple children. In contrast, early industrialized counties in central and eastern regions showed strong negative correlations, reflecting low fertility and rapid aging.

By 2010, the strong positive correlations in these regions contracted significantly, with Hunan and Jiangxi counties showing weak positive or negative correlations. Meanwhile, strong correlations shifted northward to Hubei’s counties. During this period, rural labor outmigration left a large elderly population and low birth rates.

By 2020, strong negative correlations appeared in industrial-declining counties, while areas near provincial capitals and transport hubs showed weak positive correlations. Some agricultural and mountainous counties shifted to a strong positive correlation, likely due to fertility support under rural revitalization policies.

(3)

Comparative Insights and Implications

From 2000 to 2020, central China’s core region underwent three accelerated demographic stages—high-growth mitigation, declining growth, and deep aging—5–8 years ahead of projections. Core cities (Wuhan, Nanchang, Changsha) alleviated aging through migration inflows, while peripheral, mountainous, and resource-dependent areas suffered declining births and rapid aging. County-level changes were earlier and more volatile than stable prefecture-level trends. Economic and urban development critically influenced fertility; high living costs in developed areas discouraged childbearing, lowering natural growth. Eroding traditional fertility norms, changing values, and persistent one-child policy effects further hindered short-term recovery despite two/three-child incentives.

4.2.2. Migration Rate and Population Aging

(1)

At the prefecture level (Figure 7A)

In 2000, Nanchang, Wuhan, and Changde showed strong positive correlations between outmigration and aging, driven by improved economies attracting migrants and the return of older residents. In contrast, cities like Fuzhou, Xiangyang, and Huangshi showed strong negative correlations, indicating stronger youth retention. By 2010, weak negative correlations became more common, while positive correlations clustered in the west, reflecting increased youth attraction to central cities. However, inflows were often elderly or returning residents, accelerating local aging. Regions with poor economies and weak transport saw limited migration and weak correlations, exposing development gaps. In 2020, cities such as Loudi, Xianning, and Huangshi maintained strong negative correlations, showing youth attraction. Meanwhile, cities like Yichang, Wuhan, and Xiangtan showed strong positive correlations, reflecting elderly in-migration and youth outflows. In some central cities, the migration’s impact on aging weakened, shifting to weak negative correlations, indicating subtle changes in migrant composition.

(2)

At the district and county level (Figure 7B)

In 2000, Strong positive correlations appeared sporadically, including in central, western, and southern counties, reflecting unstable migration patterns and limited economic polarization. By 2010, counties near cities like Jingmen and Xiangyang showed both strong negative and positive correlations, while most Hunan and Jiangxi counties had weak correlations, indicating stable or unclear migration-aging dynamics. By 2020, strong positive correlations expanded, especially in peripheral counties (e.g., Baokang, Yanling, Yushan) and those near provincial capitals (e.g., Tuanfeng, Nanchang, Xiangtan). This trend was prominent in economically underdeveloped and labor-exporting counties, driven by elderly return migration, fewer young migrants, and natural aging.

(3)

Multi-Scale Insights and Underlying Mechanisms

The study reveals distinct aging patterns across administrative levels: urban centers experience slower aging due to population concentration, while peripheral areas face accelerated aging from outmigration. Prefecture-level aging remains stable, whereas county-level dynamics show greater fluctuations, with cyclical return-and-outmigration patterns. Key drivers include economic polarization attracting youth to core cities, transport improvements transforming rural areas into labor-export zones, and varying elderly care traditions prompting return migration or family reunification. New urbanization’s late start hinders youth retention, and recent policies have not reversed trends. Multi-level governance is proposed: prefectures should coordinate metropolitan development and industrial relocation, while counties need tailored strategies for peripheral infrastructure and specialty industries.

4.2.3. Per Capita GDP and Population Aging

(1)

At the prefecture level (Figure 8A)

Most cities in 2000 showed a positive correlation. Cities such as Wuhan, Yichang, Changde, and Yiyang exhibited strong positive correlations, as they were undergoing rapid industrialization driven by labor-intensive industries and demographic dividends—for example, Wuhan’s manufacturing, Yichang’s petrochemical sector, and agricultural processing in Changde and Yiyang—which attracted large numbers of rural young and middle-aged laborers. In contrast, cities like Xianning, Huangshi, and Huanggang exhibited strong negative correlations, reflecting strong economic vitality and a high concentration of young laborers.

By 2010, areas with strong positive correlations had contracted, remaining in regions such as western Hunan and southern Hubei. These areas continued to rely on low factor costs to host labor-intensive industries, with limited industrial upgrading and a continued dependence on young labor. Only Qianjiang showed a strong negative correlation, as the upgrading of its petrochemical and agricultural industries provided stable employment, attracting and retaining younger workers.

By 2020, regional disparities had further intensified. Cities such as Xianning, Nanchang, and Ganzhou showed significant negative correlations (in blue), driven by the successful development of high-tech industries in Nanchang, electronics and furniture manufacturing in Ganzhou, and cultural tourism and healthcare in Xianning.

(2)

At the district and county level (Figure 8B)

In 2000, the rapid expansion of provincial capital cities 2000 initially stimulated industrial development in adjacent suburban counties, including development zones and industrial parks. These areas showed strong positive correlations, while urban districts of some prefecture-level cities exhibited strong negative correlations. This suggests that, at a time when aging pressure was not yet widespread, the effects of economic development on aging were concentrated in core urban areas.

A turning point emerged in 2010, with negative correlation zones appearing in some highly industrialized areas (e.g., counties within Xiangyang, Yichang, and Yueyang). In addition, strong positive correlations emerged in some resource-dependent cities, such as areas around Xiangtan and Tianmen, where resource depletion or rising extraction costs led to industrial decline, economic downturns, and large-scale outmigration of young workers.

By 2020, the differentiation pattern became more evident. Positive correlation zones were scattered yet widespread, indicating that aging pressure had spread from core suburban areas to economically lagging counties with limited industrial diversity and continued population outflow. In contrast, negative correlation areas were mainly located in urban fringe zones, where industries—especially manufacturing—relocated to lower-cost counties, creating new growth centers and job opportunities.

(3)

Multi-Scale Insights and Underlying Mechanisms

The correlation patterns observed at both the prefecture and district-county levels jointly reveal that in the rapid industrialization and urbanization of the Triangle of Central China, several deep-rooted drivers—including the success or failure of industrial restructuring, the direction and intensity of population mobility (especially among the young and working-age population), and the spatial effects of infrastructure (particularly transportation) reconfiguration—have fundamentally shaped the spatial divergence between aging and economic development. Core cities, through successful industrial upgrading, have managed to attract younger talent, resulting in a “wealth with youth” pattern. In contrast, traditional industrial zones, resource-depleted regions, and vast counties lacking industrial competitiveness have fallen into the dilemma of “aging before affluence” or “growth with aging”. The improvement of transportation networks has enhanced the agglomeration effect of core cities while reshaping development potential in peripheral areas. Moreover, the stark disparities in social security and public services between regions, along with the spatial imbalance and time-lag effects of new urbanization policies, have further compounded the complexity of this spatial differentiation.

4.2.4. Illiteracy Rate and Population Aging

(1)

At the prefecture level (Figure 9A)

In 2000, western regions of Hunan and Hubei, such as Changde and Yiyang, exhibited strong positive correlations between illiteracy and aging, due to the accumulation of historically high illiteracy rates and a rapidly aging population. The main causes were economic underdevelopment and a lack of educational resources. In contrast, cities in central and eastern regions such as Yueyang and Huangshi showed negative correlations, as the influx of low-educated young workers reduced the overall aging rate.

By 2010, the weakening of positive correlations was mainly driven by policy interventions and structural adjustments resulting from the return of migrant workers. National compulsory education campaigns, such as the “Two Basics” initiative, and literacy programs expanded to western regions, reducing the size of the illiterate population of older adults in cities like Changde and Yichang. Accelerated urbanization further promoted rural-to-urban migration, indirectly raising the overall level of education.

By 2020, the Prefecture-level pattern was dominated by negative correlations with localized rebounds. This was rooted in the uneven distribution of educational resources and demographic pressures at different development stages. Cities with strong educational infrastructure, such as Wuhan and Xiangyang, showed significant negative correlations, reflecting their sustained ability to attract highly educated youth and alleviate aging pressures. In contrast, traditional agricultural areas such as Tianmen and Huanggang experienced renewed positive correlations due to the continued outflow of quality educational resources and limited improvements in elderly education.

(2)

At the district and county level (Figure 9B)

In 2000, most areas with strong positive correlations were concentrated in remote or mountainous counties, such as those in southern-central Hunan, southwestern Hubei, and northwestern Jiangxi, indicating that the elderly population in these regions had generally low levels of education. In contrast, counties with convenient transportation along the Yangtze River and those surrounding provincial capitals (e.g., the Wuhan metropolitan area) showed strong negative correlations, reflecting how concentrated educational resources and an influx of younger populations helped mitigate aging.

By 2010, with the expansion of compulsory education and intensified literacy efforts, overall correlations weakened. However, certain areas—such as Zhongxiang, Tianmen, and Xiangtan County—still showed high regression values, largely due to geographic constraints, weak economies, and limited educational investment.

By 2020, with intensified aging and unequal rural educational resources, counties adjacent to urban cores—such as Taoyuan, Xiangtan, and Xiantao—again exhibited significant positive correlations. This indicates that disparities in basic education remain a key factor in localized aging. Nonetheless, a few negative correlations persisted in urban centers, suggesting that educational advantages in city cores have begun to diffuse outward.

(3)

Multi-Scale Comparison and Implications for Policy

The prefecture-level and county-level analyses are complementary in revealing the relationship between aging and illiteracy. At the prefecture level, the shift from red to blue clearly demonstrates the overall progress in educational attainment within the Middle Yangtze River Urban Agglomeration. This trend reflects a gradual weakening of the influence of illiteracy rates on aging, attributable to macro-level policies and investments in urban educational resources.

Conversely, the district and county level analysis identifies structurally disadvantaged areas masked at the broader prefecture level, such as rural or mountainous regions surrounding some cities. These areas, characterized by scarce educational resources and significant outmigration, persistently exhibit a strong positive correlation. This spatial differentiation, characterized by “urban blue” and “peripheral red” zones, indicates that the district and county levels are more suitable for formulating targeted educational policies. Furthermore, changes observed at the prefecture level are relatively stable, indicating policy continuity. In contrast, district and county-level data show more pronounced fluctuations, revealing instability in policy implementation and lags.

As cohorts born in the 1980s and 1990s enter middle and old age, illiteracy rates are expected to continue declining in the coming decades. The correlation between illiteracy and aging will likely shift from positive to negative, especially in urban areas with better access to education. However, significant regional disparities exist: provincial capitals and regions with stronger educational foundations already exhibit weak or negative correlations, indicating that educational improvements effectively weaken the link between aging and illiteracy. In contrast, counties in central and western China (such as Taoyuan, Xiantao, and Xiangtan) maintain a strong positive correlation. This reflects the dual pressures of inadequate basic education and youth outmigration in these areas, resulting in educational improvements lagging behind the pace of population aging.

4.2.5. Per Capita Number of Hospital Beds and Population Aging

(1)

At the prefecture level (Figure 10A)

In 2000, population mobility was accelerating but had not yet peaked. The spatial distribution of healthcare resources still reflected the legacy of the planned economy and was strongly tied to local demographic structures. Overall, the correlation between per capita hospital beds and aging rates was positive. In southwestern cities such as Yichang, Changde, and Yiyang, healthcare resources were concentrated, but economic development was limited. These cities experienced a significant outmigration of the working-age population, leaving behind large numbers of stay-behind and returning elderly residents, resulting in a strong positive correlation between hospital beds and aging rates. Conversely, cities such as Xianning and Huangshi, which had early foundations in industrialization, concentrated medical resources in urban cores and maintained relatively youthful populations, resulting in a strong negative correlation.

By 2010, accelerated urbanization had reshaped the spatial pattern. Core cities such as Changsha and Xiangyang attracted younger populations through economic agglomeration effects. However, hospital bed capacity could not keep pace with the aging trend or had reached saturation, leading to a slowdown in growth. Cities such as Yichang and Changde continued to show strong positive correlations, reflecting an inertia in the spatial coupling between healthcare infrastructure and aging. During this period, healthcare expansion increasingly catered to younger populations, with the construction of tertiary hospitals and the development of specialized care systems.

By 2020, more cities showed weak correlations at the prefecture level. While this might superficially suggest a balanced allocation of resources, it in fact reflects increasing polarization. Some cities, including Yichang, Changde, and Pingxiang, exhibited strong positive correlations due to the proactive expansion of healthcare infrastructure (e.g., tertiary hospital upgrades) and continued attraction of younger labor forces. This masks aging problems in peripheral areas. New strong negative correlations emerged in central cities such as Xiangyang, Wuhan, Nanchang, and Jingdezhen, where healthcare development and demographic rejuvenation were more synchronized. Meanwhile, peripheral cities like Pingxiang and Yiyang remained trapped in persistent aging with limited healthcare improvements. Some cities, such as Jingdezhen, attracted younger workers through industrial transformation while expanding healthcare infrastructure. These patterns reflect a regional restructuring of competitiveness and the influence of public health policy bias toward core cities.

(2)

At the district and county level (Figure 10B)

In 2000, areas with a “strong positive correlation” were widely distributed, covering central Hunan, northern Jiangxi, and parts of eastern and western Hubei. The outmigration of the working-age population led to a surge in left-behind elderly residents, resulting in a sharp mismatch between the demand and supply of hospital beds.

By 2010, the spatial structure had changed significantly, showing a trend of concentrated contraction. Most areas exhibited a “weak positive or negative correlation,” with only a few regions—such as western Hubei—showing a “strong positive or negative correlation.” The development of township hospitals partially alleviated medical shortages. However, intensified population mobility caused dramatic demographic shifts in some counties, leading to a mix of positive and negative correlations. In remote mountainous areas, although medical facility infrastructure improved, a shortage of healthcare professionals persisted, resulting in underutilized resources.

By 2020, the overall spatial pattern had become more complex, with red and blue zones interwoven. In underdeveloped areas, the contradiction between aging pressure and insufficient medical resources was particularly pronounced, with regions such as southwestern Hubei, northern Jiangxi, and central Hunan exhibiting a strong positive correlation. Meanwhile, district and county areas surrounding core cities, benefiting from industrial transfer and an influx of young families, formed zones of negative correlation. These areas experienced a positive cycle of improved medical resources and population rejuvenation.

(3)

Comparative Insights and Implications

At the prefecture level, core cities generally had sufficient healthcare resources and lower levels of aging, resulting in weak or negative correlations—a pattern of “abundant resources, low demand.” In contrast, peripheral cities with high aging ratios and resource shortages showed “strained supply” and strong positive correlations. In comparison, the district and county-level analyses more clearly revealed the direct relationship between aging and healthcare resources, along with greater spatial heterogeneity. Between 2000 and 2020, the expansion of positively correlated areas occurred more rapidly at the county level than at the prefecture level, suggesting that the healthcare burden brought by aging is more acute and direct at the grassroots level.

5. Discussion

5.1. Structural Differences and Complementarity from a Multi-Scale Perspective

Research shows that the prefecture-level and district and county-level scales exhibit distinct complementary characteristics in reflecting the spatial evolution of aging. The prefecture-level scale is more suitable for identifying regional collaborative development and policy effectiveness. Core cities, such as Wuhan, Changsha, and Nanchang, demonstrated sustained aging control ability from 2000 to 2020, with a relatively slow aging growth rate. This finding is consistent with the conclusions of Miao Fu [16] regarding the structural impacts within the Triangle of Central China.

The county-level scale, on the other hand, reveals the internal heterogeneity of the city and can more accurately identify the evolution paths of structural weaknesses and spatial agglomeration. For example, some peripheral counties around the cities, such as the southwestern mountainous areas of Hubei, western Hunan, and northern Jiangxi, remain in a state of deep aging, highlighting greater challenges in grassroots governance. In addition, the spatial clustering of aging shows a fluctuating trend of high elderly concentration from south to north at the prefecture level, while at the district and county level, it evolves from a “point-like” to a more “area-wide” clustering pattern. This result suggests that the prefecture-level scale is suitable for identifying the overall trends in regional collaborative development and policy impact, while the county-level scale is more helpful in capturing structural weaknesses within the city. In future regional governance, attention should be given to integrating both scales to enhance the breadth and depth of policy coverage. This phenomenon echoes the “internal structural differentiation” observed by Wu Lingguo et al. [14] in their county-level study of the three northeastern provinces.

5.2. Spatial Heterogeneity and Coupled Mechanisms of Influencing Factors

The integration of the Geographical Detector and the Geographically Weighted Regression (GWR) model effectively uncovers the spatial distribution patterns and underlying mechanisms of population aging. The Geographical Detector quantifies the global explanatory power of different influencing factors on aging, whereas the GWR model reveals spatial-scale variations in the dominance of these factors.

GWR analysis indicates significant spatial heterogeneity in the drivers of aging across different regions. For example, the impact of per capita GDP on aging displays opposite directions between core cities and their surrounding areas, reflecting a spatial coexistence of both “wealth and aging” and “wealth and youth.” This pattern has also been observed in studies of Shenzhen (Wang et al., 2017) [42]. The core mechanism lies in the ability of economically strong cities to attract younger migrants, thereby alleviating local aging, whereas adjacent industrial or resource-dependent cities, despite high GDP levels, fail to generate a comparable population absorption effect, resulting in accelerated aging due to youth outmigration.

In addition, urban-rural structural disparities are critical in shaping these mechanisms. The number of hospital beds, for example, demonstrates limited explanatory power at the county level, suggesting a structural imbalance of resource concentration at higher administrative levels and shortages at the grassroots. The correlation between illiteracy rates and aging has shifted from positive to negative, indicating that generational transitions in education are weakening the traditional link between low education and high aging. Similar trends have been reported in the Tokyo metropolitan area and Seoul (Leyso et al., 2023; Hey & Kim, 2020) [13,23], although such transformation is still in its early stages in underdeveloped regions of China.

5.3. Regional Specificity and Commonality

Compared with China’s eastern coastal areas, the Triangle of Central China exhibits both similarities and differences in aging dynamics. In terms of commonalities, both regions demonstrate a spatial divergence: aging is mitigated in urban cores while intensified in rural outmigration zones. However, due to the less pronounced economic gradient in the central region, the population siphoning effect of core cities on surrounding counties is relatively weak. This has led to the emergence of “high-aging peripheral zones,” particularly in northern Jiangxi and western Hunan. Compared to the Yellow River Basin and Northeast China, the central triangle has not yet fully entered a stage of comprehensive deep aging. Instead, it still shows a “governable window period,” offering a strategic opportunity to develop coordinated regional governance mechanisms.

Based on these findings, policies at the prefecture level should prioritize coordinated regional development, foster the “silver economy,” and strengthen the resource spillover effects of core cities. At the district and county level, efforts should aim to enhance educational and healthcare coverage, curb population outflow, and cultivate local industries to encourage youth return.

5.4. Research Contributions and Methodological Reflections

By employing a “Geodetector + GWR” methodological framework, this study balances global explanatory power with local spatial variation, offering a more comprehensive approach to capturing the spatial mechanisms of aging. Compared to traditional models such as OLS or Principal Component Analysis (PCA), this method better addresses the issues of spatial heterogeneity and factor interaction inherent in regional aging processes. However, it should be noted that the current variable selection mainly relies on static indicators. Future research may benefit from incorporating population forecasting models (e.g., age-structure simulation or cohort-component methods) and panel data approaches to better characterize the temporal evolution and path dependency of aging dynamics.

6. Conclusions and Recommendations

6.1. Conclusions

(1)

Spatiotemporal Evolution Pattern: Accelerated Aging Process and Emerging Spatial Distribution

From 2000 to 2020, the Triangle of Central China experienced an intensifying population aging problem, characterized by clear temporal stages and spatial dynamics. Data from both prefecture level and district and county levels show a significant increase in aging levels over the 20 years, with a sharp rise in the proportion of severely aged areas—over 45% of counties reached severe aging status by 2020. The spatial distribution of aging in prefecture-level cities and counties exhibited similar patterns of spatiotemporal evolution. The aging pattern evolved from an early “scattered point” distribution to a more “clustered area” pattern, forming a “high in the northwest, low in the southeast” trend, with notably high concentrations in western Hubei and the Wuhan metropolitan area.

(2)

Spatial Differentiation: Pronounced Polarization and Increasing Core-Periphery Disparities

Between 2000 and 2020, population aging in the Triangle of Central China showed an increasing trend of spatial concentration at both city and district, and county levels. High-aging areas expanded northward, with deepening aging levels and increasingly concentrated regional distribution. In contrast, low-aging areas were mainly concentrated in Jiangxi Province, forming a relatively continuous geographic pattern. The spatial pattern of aging shifted from a “scattered point” distribution to a “linear concentration.” Core urban areas such as Wuhan and Nanchang experienced relatively slow aging, while surrounding counties in western Hubei, western Jiangxi, and southern Hunan witnessed rapid aging.

(3)

Analysis of Influencing Factors: Multidimensional Drivers and Spatial Heterogeneity

The prefecture-level perspective reveals the overall regional development trends. In economically developed, highly educated, and urbanized cities such as Wuhan and Nanchang, the impact of aging on social development is relatively minor. This suggests that, supported by long-term urbanization and education policies, large cities possess stronger resource capacity and social security systems, which help mitigate the pressures of population aging. In contrast, the county-level perspective reflects more localized and detailed issues. In mountainous and impoverished counties with complex terrain and limited educational resources, aging continues to significantly raise illiteracy rates. In these regions, elderly populations tend to have low education levels and limited access to public resources, intensifying the impact of aging.

6.2. Policy Recommendations

At the prefecture level, emphasis should be placed on overall coordination and the functional differentiation of regions. For core cities such as Wuhan, Changsha, and Nanchang, it is recommended that they take the lead in developing a comprehensive elderly care service system. Given their advantages in healthcare resources, educational standards, and industrial foundations, these cities should be encouraged to promote the “silver economy.” This includes the development of high-end senior living communities, elderly care technology parks, and the advancement of smart healthcare and remote care services. In addition, industrial transfer and resource-sharing mechanisms should be utilized to support surrounding cities facing severe aging challenges, thereby alleviating regional pressure. For peripheral cities with higher levels of aging—such as those in western Hubei, northwestern Hunan, and northern Jiangxi—greater financial and policy support is necessary to advance elderly care governance. Key measures include strengthening the construction of elderly care institutions, optimizing the allocation of medical resources, and establishing senior service centers to enhance public service delivery. It is recommended to establish an intercity collaboration mechanism to construct a regional elderly care service network, enabling the sharing and flow of healthcare, education, and elderly care resources. Moreover, a population aging monitoring and early-warning platform should be developed at the prefecture level to dynamically track demographic changes and ensure precise policy implementation and resource allocation. Finally, talent mobility policies should be improved to encourage young people to remain in cities for employment and settlement, thereby mitigating the accelerating trend of population aging.

At the district and county level, policies should focus on “targeted interventions,” tailored to local realities. For mountainous and impoverished counties with high aging levels and scarce resources—such as areas in western Hubei, southwestern Hunan, and northwestern Jiangxi—investment in infrastructure should be increased. Priority should be given to building local nursing homes, daytime care centers, and wellness stations. In remote regions, the implementation of “health stations” and mobile doctor programs is advised to ensure basic medical services for older adults. Local development of elderly care industries—such as ecological agriculture and wellness tourism—should be encouraged to attract young people to return home for employment and entrepreneurship, thereby reducing outmigration and easing aging pressures. In districts with relatively low current aging levels but high future risk (e.g., some parts of the Yangtze River Economic Belt and development zones), a proactive layout of elderly care service networks is essential. This includes the establishment of smart elderly care demonstration communities and embedded community-based care centers. For isolated “aging islands”—counties experiencing deep aging in isolation—it is recommended to establish “relocated elderly care zones,” where nearby economically stronger areas provide support services to realize regional resource complementation. Additionally, a dynamic management system for population aging should be created at the district and county levels to regularly release demographic and risk assessment reports, ensuring early detection and intervention. For counties with high illiteracy rates and insufficient educational resources, priority should be given to building senior education and vocational training centers to cultivate urgently needed talent in nursing and rehabilitation, thereby alleviating service capacity shortages. These measures aim to improve and expand grassroots elderly care services, ensuring that older adults can access basic care and medical services “at their doorstep.”