Risk-adjusted capitation payment for outpatient services in China's healthcare insurance: a case study of an affluent city in southern China

Introduction This study explores the feasibility of a risk-adjusted outpatient capitation payment model for China's Urban Employee Basic Healthcare Insurance (UEBHI) and Urban-Rural Resident Basic Healthcare Insurance (URRBHI), using 2023–2024 medical insurance settlement data from GZ, an economically developed city in southern China. Methods Through multiple linear regression analysis, gender, age, Class A Outpatient Chronic and Special Diseases (OCSDs) status, and inpatient status were identified as key risk-adjustment factors ( p < 0.001), while length of hospital stay was excluded due to its lack of statistical significance in the UEBHI model ( p = 0.917). K-means clustering was applied to optimize groupings: age was categorized into 3 groups for both insurance schemes, and single Class A OCSDs were divided into 4 groups (for UEBHI) and 5 groups (for URRBHI), with low intra-group coefficients of variation (CV) ensuring grouping precision. Results A multi-year weighted adjustment mechanism was introduced to smooth fluctuations from single-year data. When equal weights were assigned to 2023 and 2024 data, the model yielded a simulated municipal healthcare insurance fund saving potential of 641 million RMB, accounting for 5.47% of the actual total outpatient fund expenditure in 2024. It should be clarified that this saving potential is a model-based estimate derived from the simulation, not an observed outcome after real-world reform implementation. Discussion This study provides a referenceable technical framework for China's outpatient healthcare insurance payment reform in economically developed regions of China, while acknowledging limitations such as the single-case sample (GZ only) and the absence of socioeconomic and lifestyle factors in the model. Future research could further refine the model by incorporating additional influencing factors and expanding the sample scope.