健康保險是保戶發生疾病或意外事故時填補其經濟損失的機制，於理賠申請時最能顯示其重要性，並可檢視保險公司履行承諾的誠信。在台灣自1995年實施全民健保以來，健康保險的需求不減反增；加上醫療產業發展迅速，更助長了保險的需求。保險公司長期穩健的經營在老年化與少子化的趨勢下，是所有保戶共同的期待。 本研究主題在於探討有無理賠者之人口變項之差異、有理賠者在事故原因及求診醫院等級之差異、不同事故原因與醫院等級交互作用下，在理賠結果之影響、及有理賠者人口變項在理賠結果之差異。本研究取用個案公司健康保險之理賠資料，進行探索性研究，期透過本研究，了解影響健康保險理賠結果之因素，以提供保險公司滿足保戶的需求與遏止道德風險之參考。 本研究經過卡方檢定、集群分析與變異數分析後，有以下之發現： 1.不同的人口變項在有無理賠申請上有顯著差異。 2.不同的人口變項在不同事故原因的理賠申請上有顯著差異。 3.不同的人口變項在不同求診醫院等級的理賠申請有顯著差異。 4.不同的人口變項在理賠結果有顯著差異。 5.不同事故原因、求診醫院等級在理賠結果有顯著差異。 根據分析結果，可以了解醫務管理與健康保險的密切關係、不同因素對理賠結果之影響，足供保險公司進行差異化管理，並可為產險公司經營健康保險之借鏡。並對保險公司提出持續投資於人、健全行政作業流程及應提供醫療照護服務等相關建議。 The health insurance is the mechanism of reimbursement for the economic losses while claim event is happened, it also shows importance most while settling a claim and applying, and the sincerity that the insurance company meets commitment is looked over. In Taiwan, National Health Insurance scheme was implemented since 1995, the demand of the health insurance has not declined. Moreover, the medical industry develops rapidly; it enlarged the demand of the insurance even more. This research emphasizes at difference of settling a claim or not, event cause, hospital level and the relationship between their interaction effects. This research takes the claims settlement materials of health insurance of a case company, carry on exploratory research. Hope has a better understanding about influences of the factor of the claims settlement result and fraud prevention through this research. There are the following findings in this research: 1. Demography shows the difference in settling a claim or not. 2. Demography shows the difference in settling a claim with different event cause. 3. Demography shows the difference in settling a claim with different hospital level. 4. Demography shows the difference in claim result. 5. Event cause, hospital level show the difference in claim result. It provides insurance company to differentiate their customers based on the analysis result. Invest on employees, optimize operation process and create healthcare service are suggested strongly.
