The impact of the global budget payment on healthcare fund expenditure in integrated delivery systems: evidence from China's county-level medical alliance reform.

BACKGROUND: The global budget payment is a key policy tool for incentivizing integrated healthcare delivery systems to pursue financial sustainability. However, there is little empirical evidence showing its effects. This study evaluates the impact of the health insurance global budget payment reform implemented under China's county-level integrated healthcare consortium model on the efficiency of health insurance fund expenditure. The study also examines the heterogeneous effects across different service types. METHODS: The study used panel data from L County, and constructed a quasi-natural experiment framework. The dependent variables: total pooled fund expenditure, general outpatient fund expenditure, outpatient fund expenditure for chronic/special diseases, and inpatient fund expenditure. The independent variable was a policy dummy variable representing the implementation of the consortium-based global budget payment reform. A two-way fixed effects model controlled for unobserved county-level heterogeneity and time trends. Potential endogeneity and model robustness were assessed using a random effects model with the Hausman-Taylor estimator and a series of robustness checks. RESULTS: Baseline regressions indicate that policy implementation is significantly negatively correlated with total medical insurance expenditures (H1) and inpatient expenditures (H4) (p < 0.01). No significant effect is seen for outpatient (H2) or chronic and special disease expenditures (H3). The fixed-effects model further confirms the negative effects on H1 and H4 (p < 0.01) and shows a significant positive association with outpatient expenditures (p < 0.05). This indicates potential cost-shifting effects. Both the Breusch-Pagan test (p < 0.05) and the Hausman test support the adoption of the fixed-effects model. The results remain robust in the two-way fixed-effects specification. Model explanatory power varies: total and inpatient expenditure models show relatively high explanatory capacity (R 2 = 0.273-0.538), while the chronic and special disease model shows a weak fit (R 2 = 0.031). CONCLUSION: Within China's county-level integrated healthcare consortium framework, the health insurance global budget payment reform effectively curbs the overutilization and growth in costs of inpatient services, but may induce a shift in services towards outpatient care. This cost-shifting effect indicates that isolated inpatient global budgeting, without a coordinated payment mechanism design for outpatient services, may not optimize the overall expenditure efficiency of health insurance funds.