OBJECTIVES: The diagnosis-related group (DRG) based reimbursement system has been voluntarily applied to inpatients with seven diseases in the Korean national health insurance since 2002, and was mandatory for all health-care institutions from July 1, 2013. The main purpose of this study was to evaluate the performance of DRG-based reimbursement in health care expenditure and to propose alternative policies. METHODS: A non-equivalent control group pretest-posttest design with a difference-in-difference approach was adopted to compare changes in medical service utilization and physician's behavior between DRG-based reimbursement(experimental group) and fee-for-service reimbursement(control group). Seven diseases to which DRG-based reimbursement was applied included tonsillectomy, cataract surgery, appendectomy, herniotomy, hemorrhoidectomy, hysterectomy, and Caesarean section. The panel data were produced from year 2004~2011 medical claims database of the National Health Insurance, which covered a total of 1,119,028 cases per year. RESULTS: From 2004 to 2011, surgical operations in institution reimbursed by DRG have been significantly increased more than those in institutions reimbursed by fee-for-service. The results showed that the DRG-based payment has reduced the length of stay in seven diseases, while it has changed physician’s behavior to charge DRG-code upward and shift medical tests and expensive antibiotics from inpatients to outpatients because DRG was applied to inpatient only. The DRG-based payment in seven diseases has consistently increased medical expenditure as well as medication expenses more than fee-for-service, partly due to no global budget in the Korean national health insurance. CONCLUSIONS: Challenges and future issues to expand the DRG-based reimbursement system to all diseases for inpatients should be considered such as monitoring service quality, strategic plans to control physicians’ behavior, limiting the number of DRG classifications, and the introduction of global budgeting.