Impact of outpatient pooling reform on inpatient expenditure and quality of care: evidence from an interrupted time series analysis in China

Abstract

Public health is a cornerstone of sustainable development; however, rapidly rising healthcare expenditure has placed increasing pressure on health insurance systems worldwide. Although China has achieved near-universal coverage through its basic medical insurance system, it has long faced a structural imbalance characterized by a preference for inpatient over outpatient services. The outpatient pooling reform under the Urban Employee Basic Medical Insurance (UEBMI) aims to promote a more rational distribution of healthcare utilization, optimize resource allocation, and improve the efficiency of insurance fund use. This study examines changes in inpatient expenditure and quality of care following the implementation of this reform. We used micro-level claims data for urban employee inpatients from a province in China, covering the period from January 2022 to December 2024. An interrupted time series analysis (ITSA) design was employed to assess changes in trends of inpatient expenditure and quality-of-care indicators before and after the implementation of the outpatient pooling reform. Although an immediate increase in the monthly mean expenditure per inpatient admission and several itemized expenditure categories was observed at the time of policy implementation, the post-reform monthly trend decreased significantly relative to the pre-reform trajectory. Specifically, monthly mean expenditure per inpatient admission declined by approximately 45 yuan per month during the post-reform period ( \(P < 0.001\) ). Both health insurance reimbursement and out-of-pocket payments (OOP) also showed significant downward post-reform trends (both \(P < 0.001\) ). Pre-policy rising trends in drug, examination, treatment, and bed expenditure were effectively curbed after the reform. Regarding selected care-related indicators, the average length of stay (LOS) showed a decreasing post-reform trend. The 30-day readmission rate showed a small statistically significant immediate increase at policy implementation ( \(\beta_2 = 0.016\) , \(P = 0.048\) ), whereas neither the subsequent slope change nor the net post-reform trend was statistically significant. Heterogeneity analysis indicated that the estimated magnitude of changes varied across hospital levels, with larger point estimates observed in higher-level hospitals and relatively limited changes in primary hospitals. The UEBMI general outpatient pooling reform was associated with sustained reductions in monthly mean expenditure per inpatient admission following an initial increase at policy implementation. These changes may have contributed to easing the financial burden on insured inpatients, while favorable trends were also observed in selected process-of-care indicators, particularly LOS. The estimated policy-associated changes varied across hospital levels. Future reforms should strengthen the capacity of primary healthcare institutions and further advance payment system reforms to enhance policy effectiveness.