An Actuarial Model for Hospitalization Insurance with Limited Benefit: Cancer Impaired Risk

• Main Authors: Chang-Li Lin, 林長立
• Other Authors: Shing-Her Juang
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/24893573970997783185

碩士 === 東吳大學 === 財務工程與精算數學系 === 101 === The demand for private medical insurance is increasing in Taiwan as public awareness about healthcare needs. Some of the most popular individual hospitalization insurance plans are limited and fixed premium payment products, with unlimited or lifetime benefits. With the average lifespan increasing and the rapidly rising cost of healthcare, a lifetime benefit creates substantial risk to health insurance providers. Insurance companies have stopped offering unlimited and lifetime benefits. The two types of insurance contracts currently offered either provides limited benefits or the unused portion as a life insurance benefit. Hospitalization benefits can be classified as actual daily medical expense, fixed daily benefit, or the greater of the two. In this study, we consider a hospitalization insurance for newly diagnosed cancer patients with a fixed daily benefit and lifetime claims, with ceilings. We use a generalized linear model (GLM) to analyze thirteen-year inpatient longitudinal data of 695,978 cancer patients, diagnosed between 1997 and 2009, which is approximately a third of Taiwan’s new cancer cases in that period. Data are extracted from Taiwan’s National Health Insurance Research Database via ICD_9_CM codes. Our study indicates that the special GLM model yields reasonable results for annual inpatient days, to be specific, the generalized estimating equation, negative binomial distribution, and exchangeable correlation. We then used copula and simulation methods to determine the net premium required under varied claims ceiling, mortality rates, and correlation coefficients. For premium paying period, we consider both single premium and regular premium. Consider also given to the regular premium with or without rational termination. Our findings are that firstly, there is a correlation for annual inpatient days; the higher correlation for lower premiums. This statement is true for hospitalization insurance with and without a life insurance benefit. Secondly, an increase in the mortality rate will raise the premium for the hospitalization insurance with life insurance benefit, however it will reduce the premium for hospitalization insurance without a life insurance benefits. Third, the net premium reserve depend not only the policy year but also the benefits already paid. Reserve may not be needed in the first few policy years. Fourth, premium is lower with rationality termination consideration than without rational termination.