Akihiro Yoshimura, Reo Takaku, Medium-run effects of patient cost-sharing on the demand-side and supply-side of inpatient care: A natural experiment in Japan, Japan and the World Economy, Volume 76, 2025, https://doi.org/10.1016/j.japwor.2025.101333.
Abstract:
Over the last several decades, many developed countries have increased patient cost-sharing to contain healthcare expenditures. Some influential studies, including the RAND Health Insurance Experiment, suggest that increasing patient cost-sharing reduces unnecessary medical care by mitigating ex-post moral hazard. However, most existing studies in this field focus on short-run effects, in which the health status of the insured and provider behaviors remain unchanged. In this study, we analyze the medium-term impact of the 2003 coinsurance rate increase on inpatient care utilization by using public hospital data in Japan covering a 12-year period. Difference-in-differences and event study analyses robustly showed that the number of inpatients decreased a few years after the 2003 reforms. However, inpatient costs per patient day began to increase after four years, mainly due to increased medical resources. Eventually, despite the initial reduction in the number of inpatients, we found that the effects on total inpatient costs were negligible in the medium run. These findings indicate that many existing studies may overestimate the cost-containment effects of patient cost-sharing in the medium- and long-term.
Over the last several decades, many developed countries have increased patient cost-sharing to contain healthcare expenditures. Some influential studies, including the RAND Health Insurance Experiment, suggest that increasing patient cost-sharing reduces unnecessary medical care by mitigating ex-post moral hazard. However, most existing studies in this field focus on short-run effects, in which the health status of the insured and provider behaviors remain unchanged. In this study, we analyze the medium-term impact of the 2003 coinsurance rate increase on inpatient care utilization by using public hospital data in Japan covering a 12-year period. Difference-in-differences and event study analyses robustly showed that the number of inpatients decreased a few years after the 2003 reforms. However, inpatient costs per patient day began to increase after four years, mainly due to increased medical resources. Eventually, despite the initial reduction in the number of inpatients, we found that the effects on total inpatient costs were negligible in the medium run. These findings indicate that many existing studies may overestimate the cost-containment effects of patient cost-sharing in the medium- and long-term.
Keywords:
Demand for medical care; Natural experiment; Price elasticity; Moral hazard; Physician induced demand; Cost-sharing
Demand for medical care; Natural experiment; Price elasticity; Moral hazard; Physician induced demand; Cost-sharing