Joint spatiotemporal evaluation of multiple healthcare resources: hospitals, hospital beds and physicians across 365 Chinese cities over 22 years

Xin Qi¹Mingyu Xie¹Yaqian He²Xianteng Tang³Lingfeng Liao⁴Yaling Luo⁴Kaiwei Lin³Xiang Yan⁵Xiuli Wang⁴Yuanyuan Zhu⁴Zhangying Tang³Yumeng Zhang^6*Chao Song^6*Jay Pan⁴

• ¹School of Public Health,, Xi'an Jiaotong University, Xi'an, China
• ²Department of Geosciences, University of Arkansas, Fayetteville, United States
• ³State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, School of Geoscience and Technology, Southwest Petroleum University, Chengdu, China
• ⁴HEOA–West China Health & Medical Geography Group, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
• ⁵College of Architecture and Environment, Institute of Urbanization Strategy and Architecture Research, Sichuan University, Chengdu, China
• ⁶Institute for Healthy Cities and West China Research Centre for Rural Health Development, Sichuan University, Chengdu, China

Regional disparities in healthcare resource allocation across space and time present significant challenges to the global achievement of SDG 3, SDG 10, and SDG 11. To this end, we propose a joint spatiotemporal evaluation framework to assess the synergistic efficiency of multiple healthcare resource indicators.Using China as a case study, we analyzed data from 365 cities on three key healthcare resource indicators: hospitals, hospital beds, and physicians. A composite healthcare resource score was constructed using the entropy weight method. We developed a three-dimensional joint spatiotemporal evaluation framework incorporating spatial Gini coefficient, emerging hotspot analysis, and Bayesian Spatiotemporally Varying Coefficients (BSTVC) model with Spatiotemporal Variance Partitioning Index (STVPI) to evaluate spatiotemporal equity, agglomeration and influencing factors. Individual indicators were evaluated to validate the framework's robustness.(i) Spatiotemporal description: The composite indicator, weighted by hospitals (25%), hospital beds (46%), and physicians (29%), showed only a modest increase from 2000 to 2021, with persistently lower values in western and northern regions. (ii) Common spatiotemporal equity: The spatial Gini coefficient for the composite indicator increased annually by 0.34%, mirroring trends in hospital beds (0.34%) and physicians (0.26%) but contrasting with hospitals (-0.32%). This suggested declining equity was mainly driven by hospital beds and physicians, partially offset by the more balanced distribution of hospitals. (iii) Common spatiotemporal agglomeration: Hotspot intensity for the composite indicator was lower than that for hospitals but higher than for hospital beds and physicians. Cold spots were more concentrated for the composite indicator than for any individual indicator, with less than 10% overlap across the three indicators, indicating weak regional synergy. (iv) Common spatiotemporal drivers: BSTVC and STVPI methods revealed consistent patterns of explainable percentages across four healthcare resource indicators, with population density (37.96%, 95% CI: 30.05-43.05%) and employed population density (31.63%, 30.69-33.83%) emerging as dominant common drivers, supporting unified and coordinated policy interventions.We proposed a joint spatiotemporal evaluation framework to quantify both common and differentiated allocation patterns and driving factors across multiple healthcare resource indicators, highlighting the need for type-specific, temporally responsive, and spatially adaptive interventions to support dynamic monitoring and precise regulation of regional healthcare resource allocation over the globe.