Recently, there has been a surge in cases of fraudulent claims related to health insurance due to the increase in health insurance use and the complexity of its claim processes. In order to prevent the financial leaks of health insurance by proactively detecting these fraudulent claims, it is necessary to identify fraud detection indicators, compare their relative importance, and prioritize them. Although this is a very important issue from both academic and practical points of view, there are no previous studies directly related to it. Therefore, this study attempts to analyze the relative importance of various fraud detection indicators derived from prior literature and prioritize them by applying the method of analytic hierarchy process (AHP). In addition, it further examines whether there is any significant difference in importance of each fraud detection indicator between the two health insurance related institutions(i.e., the medical institution and the Health Insurance Review & Assessment Service). The results of this study would help improve the predictive model to detect fraudulent claims of health insurance in the short term, and help prepare strategies to reduce financial leakage in the operation of health insurance in the long term. The medical institutions and the Health Insurance Review & Assessment Service may have some practical implications from this study.