Regional disparities in the allocation of rehabilitation institutions for persons with disabilities in China: a spatio-temporal and inequality decomposition analysis

Background With the global population of persons with disabilities continuing to grow, establishing an equitable and efficient rehabilitation service system has become essential for advancing health equity and social inclusion. As key service providers within this system, the spatial distribution of rehabilitation institutions critically shapes access to rehabilitation services. This study examines the regional equity and spatiotemporal evolution of rehabilitation institutions for persons with disabilities in China from 2016 to 2023, and proposes policy recommendations for optimizing resource allocation. Methods panel data from 31 provinces (autonomous regions and municipalities) in China from 2016 to 2023 were analyzed using data from the China Statistical Yearbook on the Work for Persons with Disabilities . Three quantitative approaches were employed: Health Resource Agglomeration Degree (HRAD), Dagum Gini Decomposition, and Kernel Density Estimation. HRAD assessed the alignment between institutional resources and population distribution. Dagum Gini decomposition measured overall spatial disparities and identify their sources, while kernel density estimation illustrated dynamic distribution patterns and polarization characteristics at both national and regional levels. Results From 2016 to 2023, the total number of rehabilitation institutions increased from 7,753 to 12,410 nationwide. The eastern, central, and western regions recorded growth rates of 60.10%, 50.37%, and 61.13%, respectively. However, the eastern region consistently accounted for more than half of the national total, and absolute regional disparities continuing to widen. HRAD results show a decline in the number of provinces with insufficient resource allocation, indicating improvements in overall matching efficiency, although substantial interprovincial disparities persist. The overall Gini coefficient decreased from 0.70 to 0.62, reflecting a gradual reduction in spatial inequality. Notably, the sources of inequality shifted structurally from interregional disparities to hyper-density effects. Kernel density estimates further reveal that while polarization has eased since 2020, the eastern region exhibits a trend toward medium-to-high stratification, and the western region shows emerging signs of polarization. Conclusions Although China has achieved substantial growth in rehabilitation institutions and improvements in allocation efficiency. Nevertheless, pronounced regional inequities remain, increasingly shaped by hyper-density effects and stratified spatial patterns.Strengthening institutionalized governance mechanisms is essential to promote more balanced, resilient, and sustainable development of the rehabilitation service system.